Cochrane
Library
Cochrane Database of Systematic Reviews

   
Advance care planning for haemodialysis patients (Review)

  Lim CED, Ng RWC, Cheng NCL, Cigolini M, Kwok C, Brennan F  

  Lim CED, Ng RWC, Cheng NCL, Cigolini M, Kwok C, Brennan F.  

Advance care planning for haemodialysis patients.
Cochrane Database of Systematic Reviews 2016, Issue 7. Art. No.: CD010737.
DOI: 10.1002/14651858.CD010737.pub2.

  www.cochranelibrary.com  

 
Advance care planning for haemodialysis patients (Review)
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

TABLE OF CONTENTS
HEADER......................................................................................................................................................................................................... 1
ABSTRACT..................................................................................................................................................................................................... 1
PLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2
BACKGROUND.............................................................................................................................................................................................. 3
OBJECTIVES.................................................................................................................................................................................................. 4
METHODS..................................................................................................................................................................................................... 4
RESULTS........................................................................................................................................................................................................ 5
Figure 1.................................................................................................................................................................................................. 6
Figure 2.................................................................................................................................................................................................. 7
Figure 3.................................................................................................................................................................................................. 8
DISCUSSION.................................................................................................................................................................................................. 9
AUTHORS' CONCLUSIONS........................................................................................................................................................................... 10
ACKNOWLEDGEMENTS................................................................................................................................................................................ 10
REFERENCES................................................................................................................................................................................................ 11
CHARACTERISTICS OF STUDIES.................................................................................................................................................................. 13
APPENDICES................................................................................................................................................................................................. 18
CONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 23
DECLARATIONS OF INTEREST..................................................................................................................................................................... 23
SOURCES OF SUPPORT............................................................................................................................................................................... 23
INDEX TERMS............................................................................................................................................................................................... 23

Advance care planning for haemodialysis patients (Review) i

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

[Intervention Review]

Advance care planning for haemodialysis patients

Chi Eung Danforn Lim1, Rachel WC Ng2, Nga Chong Lisa Cheng3, Maria Cigolini4, Cannas Kwok5, Frank Brennan6

1Faculty of Science, University of Technology Sydney, Earlwood, Australia. 2Faculty of Medicine, University of New South Wales,
Hurstville, Australia. 3South Western Sydney Clinical School, Faculty of Medicine, University of New South Wales, Blakehurst, Australia.
4Department of Palliative Medicine, Royal Prince Alfred Hospital, Camperdown, Australia. 5School of Nursing and Midwifery, University of
Western Sydney, Parramatta, Australia. 6Department of Renal Medicine, St George Hospital, Kogarah, Australia

Contact address: Chi Eung Danforn Lim, Faculty of Science, University of Technology Sydney, C/O Earlwood Medical Centre,, 356 Homer
Street, Earlwood, New South Wales, 2206, Australia. celim@unswalumni.com, D.LIM@UNSW.EDU.AU.

Editorial group: Cochrane Kidney and Transplant Group.

Publication status and date: New, published in Issue 7, 2016.

Citation: Lim CED, Ng RWC, Cheng NCL, Cigolini M, Kwok C, Brennan F. Advance care planning for haemodialysis patients. Cochrane
Database of Systematic Reviews 2016, Issue 7. Art. No.: CD010737. DOI: 10.1002/14651858.CD010737.pub2.

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT

Background
End-stage kidney disease (ESKD) is a chronic, debilitative and progressive illness that may need interventions such as dialysis,
transplantation, dietary and fluid restrictions. Most patients with ESKD will require renal replacement therapy, such as kidney
transplantation or maintenance dialysis. Advance care planning traditionally encompass instructions via living wills, and concern patient
preferences about interventions such as cardiopulmonary resuscitation and feeding tubes, or circumstances around assigning surrogate
decision makers. Most people undergoing haemodialysis are not aware of advance care planning and few patients formalise their wishes
as advance directives and of those who do, many do not discuss their decisions with a physician. Advance care planning involves planning
for future healthcare decisions and preferences of the patient in advance while comprehension is intact. It is an essential part of good
palliative care that likely improves the lives and deaths of haemodialysis patients.

Objectives
The objective of this review was to determine whether advance care planning in haemodialysis patients, compared with no or less
structured forms of advance care planning, can result in fewer hospital admissions or less use of treatments with life-prolonging or curative
intent, and if patient's wishes were followed at end-of-life.

Search methods
We searched the Cochrane Kidney and Transplant Specialised Register to 27 June 2016 through contact with the Information Specialist
using search terms relevant to this review. We also searched the Cumulative Index of Nursing and Allied Health Literature (CINAHL), and
Social Work Abstracts (OvidSP).

Selection criteria
All randomised controlled trials (RCTs) and quasi-RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate
medical records, date of birth or other predictable methods) looking at advance care planning versus no form of advance care planning in
haemodialysis patients was considered for inclusion without language restriction.

Data collection and analysis

Data extraction was carried out independently by two authors using standard data extraction forms. Studies reported in non-English
language journals were translated before assessment. Where more than one publication of one study exists, reports were grouped together
and the publication with the most complete data was used in the analyses. Where relevant outcomes are only published in earlier versions
these data were used. Any discrepancies between published versions were highlighted. Non-randomised controlled studies were excluded.
Advance care planning for haemodialysis patients (Review) 1
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Main results
We included two studies (three reports) that involved 337 participants which investigated advance care planning for people with ESKD.
Neither of the included studies reported outcomes relevant to this review. Study quality was assessed as suboptimal.

Authors' conclusions
We found sparse data that were assessed at suboptimal quality and therefore we were unable to formulate conclusions about whether
advance care planning can influence numbers of hospital admissions and treatment required by people with ESKD, or if patients' advance
care directives were followed at end-of-life. Further well designed and adequately powered RCTs are needed to better inform patient and
clinical decision-making about advance care planning and advance directives among people with ESKD who are undergoing dialysis.

PLAIN LANGUAGE SUMMARY

Advance care planning for haemodialysis patients

Background

People with chronic kidney disease and end-stage kidney disease (ESKD) have irreversible kidney damage and require renal replacement
therapy. In developed countries, haemodialysis has become the most common treatment for people with ESKD. Despite its life-saving
potential, haemodialysis can be a significant physical and psychological burden to patients. Advance care planning is the process of
planning for a person’s future health and personal care decisions in terms of level of healthcare and quality of life the person would want,
should for any reason, the person becomes unable to participate in decision-making. Advance care goals can change over time and advance
care planning assists in addressing these changes and readdress care goals over time. This helps to ensure that individual choices are
respected in future medical treatment in an event where the person cannot communicate or make decisions.

We wanted to find out if advance care planning can improve health outcomes among people with ESKD in terms of use of resuscitation
measures such as cardiopulmonary resuscitation, and withdrawal from dialysis. Additionally, we wanted to find out whether patient wishes
were followed in the end of life period.

Study characteristics

We searched the literature up to June 2016 and found two studies (three reports) that involved 337 patients which investigated use of
patient-centred advanced care planning (PC-ACP) and peer mentoring interventions.

Key results

Neither study addressed our questions concerning use of life-prolonging treatments such as resuscitation, death in hospital or withdrawal
from dialysis. It remains uncertain if advance care planning can improve health outcomes among ESKD patients. More research is required
to better inform use of PC-ACP for people receiving haemodialysis treatment.

Advance care planning for haemodialysis patients (Review) 2

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

BACKGROUND The aim of haemodialysis is to correct metabolic and electrolyte

disturbances seen in ESKD patients. Haemodialysis uses a dialyser
Many people with chronic kidney disease (CKD) and end-stage consisting of parallel membrane plates or hollow fibres suspended
kidney disease (ESKD) have irreversible kidney damage and in dialysis fluid (dialysate), which act as a semipermeable
require renal replacement therapy. For people with ESKD, kidney membrane, across which blood and dialysate can flow. The
transplant is the treatment of choice because compared with dialysate fluid has an electrolyte composition similar to that of
haemodialysis, transplant reduces mortality risk and improves extracellular fluid and flows in an opposite direction to the blood.
quality of life. Haemodialysis has become the default treatment Solute and water can move across the semipermeable membrane,
for people with ESKD in developed countries, and although it allowing metabolic wastes and excess body water from blood to be
sustains life, for many people it is a supportive or palliative removed and replenishes body buffers before the blood is returned
therapy (Davison 2007). Age and co-morbidities of people receiving to the patient (Mallick 1999). Haemodialysis carries risks such
haemodialysis have increased. In low-risk patients, mortality is as vascular access failure, venous occlusion, and line infections
much higher compared with age-matched controls who do not (Brown 2007).
have kidney failure; and for high-risk patients, haemodialysis may
simply prolong the process of dying (Vilar 2011). Recent reports indicate a gradual plateau of incident ESKD over
time (ANZDATA 2012). There has been an increase in dialysis
Advance care planning traditionally encompasses instructions via patients dying after withdrawal of dialysis (10% to 15% in 1990,
living wills concerning patient preferences about interventions 20% in 2004) making it the second leading cause of death following
such as cardiopulmonary resuscitation and feeding tubes, or cardiovascular disease (KDIGO 2013).
circumstances around assigning surrogate decision makers.
However, few patients formalise their wishes as advance care Description of the intervention
planning directives, and of those who do, many do not discuss
Although advance care planning for each patient is individualised,
their decisions with a physician (Weisbord 2003). In many instances,
the underlying principles of the advance care planning process
patients do not address withdrawal from dialysis because they are
are similar: the responsible physician determines the patient's
unaware that this option is available, or believe that their physician
priorities, establishes agreed levels of care, manages appropriate
would not support this decision (Davison 2007). Davison 2010 study
transition from active to palliative care, achieves symptom control
of 584 patients with stage 4 and 5 CKD reported that only 38.2% had
at end-of-life, and acts as a spokesperson for the patient if needed
completed a personal directive.
(Arulkumaran 2012). Advance care plans are reviewed constantly as
McAdoo 2012 surveyed 138 kidney patients at 10 London hospitals part of an overall dynamic program of care (Holley 2003).
and found that 28% (n = 26) of those who died in hospital
The process is changing from a document-driven, decision-
had discussed end-of-life care with medical or nursing staff in
focused event to an approach that encompasses understanding,
the year before death, and 52% (n = 49) had good symptom
reflection, communication, and discussion involving patients,
control at the time of death. The absence of end of life discussion
family or caregivers, and healthcare staff to clarify end-of-life care
does not accord with findings that people with CKD most valued
preferences (Davison 2006). Advance care planning is traditionally
being informed about prognosis and in being able to prepare for
used in the event of incapacity to make informed decisions (Simon
death (Davison 2010). Poorly documented advance care planning
2008); however change in health status over time may prompt
processes resulted in the survey by McAdoo 2012 eliciting that
change about intervention preferences during the course of care.
although few patients discussed end-of-life care with medical
Discussions therefore occur early in the disease process (Davison
teams, 43% (n = 40) discontinued dialysis, and 45% (n = 42) were
2007), and patients are encouraged to appoint a spokesperson
referred for palliative care, highlighting the need for clear and well-
if needed. The goal is that preferences are communicated to all
constructed advance care planning processes and documentation.
stakeholders and documented appropriately (Russon 2010).
Advance care planning involves planning for future healthcare
decisions and preferences of the patient in advance while How the intervention might work
comprehension is intact. It is an essential part of good palliative Aiding patients to establish options and priorities for end-of-life
care that likely improves the lives and deaths of haemodialysis care enables preferences to be maintained according to individual
patients (Davison 2006). wishes (Russon 2010). Advance care plans are often not followed by
healthcare providers, and it may not be possible to anticipate all
Description of the condition clinical contingencies for individual patients. The most significant
CKD is defined as abnormalities of kidney structure or function barrier may be lack of physician intervention to initiate and guide
(e.g. albuminuria or GFR < 60 mL/min/1.73 m2), present for advance care planning discussions (Davison 2010).
greater than three months, with implications for health (KDIGO
In some instances, continuation of treatment and admission to
2013). The incidence and prevalence of CKD varies across regions
intensive care settings may prolong death rather than sustain life.
with incidence as high as 200 cases per million per year in
A well-constructed and communicated advance care plan is likely
some countries (Levey 2012). Advanced (Stage 5) CKD (GFR <
to improve satisfaction and end-of-life care for patients and reduce
15 mL/min/1.73 m2) is known as ESKD, and is associated with psychological stress for surviving relatives (Arulkumaran 2012).
metabolic complications such as uraemia, electrolyte disorders,
and hypertension, which contribute to significant morbidity and The focus of advance care planning is about living well and ensuring
mortality. optimal care for patients near the end-of-life, and not just on
death and the decision to refuse treatment. Advance care planning
enables patients to prepare for death, strengthen relationships
Advance care planning for haemodialysis patients (Review) 3
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

with loved ones, achieve control over their lives, and relieve burden 5. Treatment goal nearing the end-of-life (life prolonging/
on others (Singer 1995). palliation)
6. Inpatient hospital admission near end-of-life (yes/no)
Why it is important to do this review
7. Length of stay in hospital (continuous)
Despite its life-saving potential, haemodialysis can impose 8. Awareness among healthcare providers of patients' stated
significant risks of morbidity and mortality, as well as a range preferences during end-of-life care and advance decisions (yes/
of physical and psychological symptoms and complications no).
(Weisbord 2003). Advance care planning may be beneficial in
providing better quality of care at end-of-life and offer more Search methods for identification of studies
autonomy in health care decision-making for people with ESKD
Electronic searches
receiving dialysis treatment.
We searched the Cochrane Kidney and Transplant Specialised
OBJECTIVES Register to 27 June 2016 through contact with the Information
Specialist using search terms relevant to this review. The Cochrane
The objective of this review was to determine whether advance Kidney and Transplant Specialised Register contains studies
care planning in haemodialysis patients, compared with no or less identified from the following sources.
structured forms of advance care planning, can result in fewer
hospital admissions or less use of treatments with life-prolonging 1. Quarterly searches of the Cochrane Central Register of
or curative intent, and if patient's wishes were followed at end-of- Controlled Trials (CENTRAL)
life. 2. Weekly searches of MEDLINE OVID SP
METHODS 3. Handsearching of kidney-related journals and the proceedings
of major kidney conferences
Criteria for considering studies for this review 4. Searching of the current year of EMBASE OVID SP
Types of studies 5. Weekly current awareness alerts for selected kidney journals
6. Searches of the International Clinical Trials Register (ICTRP)
All randomised controlled trials (RCTs) and quasi-RCTs (RCTs in Search Portal and ClinicalTrials.gov.
which allocation to treatment was obtained by alternation, use
of alternate medical records, date of birth or other predictable Studies contained in the Specialised Register are identified through
methods) looking at advance care planning versus no form of search strategies for CENTRAL, MEDLINE, and EMBASE based on
advance care planning for haemodialysis patients were included. the scope of Cochrane Kidney and Transplant. Details of these
strategies, as well as a list of handsearched journals, conference
Types of participants proceedings and current awareness alerts, are available in the
Inclusion criteria Specialised Register section of information about the Cochrane
Kidney and Transplant.
Studies eligible for inclusion involved people with ESKD
undergoing haemodialysis. We also searched the Cumulative Index of Nursing and Allied
Health Literature (CINAHL), and Social Work Abstracts (OvidSP) to
Exclusion criteria December 2014.
We excluded studies that involved people with clinically-diagnosed
See Appendix 1 for search terms used in strategies for this review.
mental illness (e.g. schizophrenia, dementia), and people with
ESKD not on haemodialysis. Searching other resources
Types of interventions 1. Reference lists of clinical practice guidelines, review articles and
relevant studies
We included studies that investigated the use of advance care
planning. 2. Letters seeking information about unpublished or incomplete
studies to investigators known to be involved in previous
Types of outcome measures studies.
Primary outcomes Data collection and analysis
1. Use of treatments with primary life-prolonging or curative intent Selection of studies
order in place, such as resuscitation (yes/no)
2. Hospital death (yes/no) The search strategy described was used to obtain titles and
abstracts of studies that may be relevant to the review. Titles
3. Dialysis discontinued before death (yes/no). and abstracts were screened independently by two authors, who
Secondary outcomes discarded studies that were not applicable; however, studies and
reviews that might include relevant data or information on studies
1. Good symptom control achieved (yes/no) were retained initially. Two authors then independently assessed
2. Use of end-of-life care pathway (yes/no) retrieved abstracts, and if necessary the full text of the studies, and
3. Patient's satisfaction with end-of-life care (yes/no) determined which satisfied the inclusion criteria.
4. Caregiver's satisfaction with end-of-life care (yes/no)

Advance care planning for haemodialysis patients (Review) 4

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Data extraction and management planned to be performed. Attrition rates, for example drop-
outs, losses to follow-up and withdrawals were planned to be
Data extraction was carried out independently by two authors using
investigated. Issues of missing data and imputation methods (e.g.
standard data extraction forms. Studies reported in non-English
last-observation-carried-forward) was critically appraised (Higgins
language journals were translated before assessment. Where more
2011).
than one publication of one study exists, reports were grouped
together and the publication with the most complete data was used Assessment of heterogeneity
in the analyses. Where relevant outcomes are only published in
earlier versions, these data were used. Any discrepancies between Where required, heterogeneity was planned to be analysed using a
published versions were highlighted. Non-randomised controlled Chi2 test on N-1 degrees of freedom, with an alpha of 0.05 used for
studies were excluded. statistical significance and with the I2 test (Higgins 2003). I2 values
of 25%, 50% and 75% correspond to low, medium and high levels
Assessment of risk of bias in included studies of heterogeneity.
The following items were independently assessed by two authors
Assessment of reporting biases
using the risk of bias assessment tool (Higgins 2011) (see Appendix
2). Where required and possible, funnel plots were to be used to assess
for the potential existence of small study bias (Higgins 2011).
• Was there adequate sequence generation (selection bias)?
• Was allocation adequately concealed (selection bias)? Data synthesis
• Was knowledge of the allocated interventions adequately We planned to pool data using the random-effects model but the
prevented during the study? fixed-effect model was also to be used to ensure robustness of the
* Participants and personnel (performance bias) model chosen and susceptibility to outliers.
* Outcome assessors (detection bias)
Subgroup analysis and investigation of heterogeneity
• Were incomplete outcome data adequately addressed (attrition
bias)? Subgroup analysis was planned to be performed to investigate
• Are reports of the study free of suggestion of selective outcome if effects varied due to advance care planning process type,
reporting (reporting bias)? differences in populations and settings, and study design
• Was the study apparently free of other problems that could put (cluster-RCTs versus parallel RCTs). Other potential sources of
it at a risk of bias? heterogeneity, such as participants' age and gender was also
planned to be analysed. Adverse effects was planned to be
Measures of treatment effect tabulated and assessed using descriptive techniques. If possible,
the risk difference with 95% CI was planned to be calculated
Dichotomous outcomes (e.g. use of treatments with primary life- for each adverse effect, either compared with no treatment (i.e.
prolonging or curative intent, hospital death, hospital admission, advance care planning) or another intervention.
dialysis discontinuation prior to death, good symptom control
achieved, satisfaction with end-of-life care, treatment goal nearing Sensitivity analysis
end-of-life (life-prolonging or palliation), and healthcare providers'
awareness of patients' stated preferences) was planned to be We planned to perform sensitivity analyses to explore the influence
expressed as risk ratio (RR) with 95% confidence intervals (CI). of the following factors on effect size.
Where continuous scales of measurement were used to assess the
• Repeating the analysis excluding unpublished studies
effects of treatment (e.g. knowledge of advance care planning,
length of hospital stay prior to death) the mean difference (MD) was • Repeating the analysis taking account of risk of bias
planned to be used, or the standardised mean difference (SMD) if • Repeating the analysis excluding any very long or large studies
different scales was used. to establish how much they dominate the results
• Repeating the analysis excluding studies using the following
Unit of analysis issues filters: diagnostic criteria, language of publication, source of
We had planned to include effect estimates from cluster-RCTs funding (industry versus other), and country.
and cross-over RCTs together with estimates from parallel RCTs
where designs were properly accounted for and reported. If effect RESULTS
estimates for cluster-RCTs and cross-over RCTs were not corrected
Description of studies
appropriately for possible unit of analysis errors, we had planned to
adjust the CIs of those estimates by reducing the size of each study For detailed descriptions, please see Characteristics of included
to its effect sample size. studies, Characteristics of excluded studies and Characteristics of
studies awaiting classification.
Dealing with missing data
Results of the search
Any further information required was sought from the original
author by written correspondence (e.g. emailing or writing to Our search identified 77 potentially relevant records; of these, 52
corresponding author) and any relevant information obtained in were duplicates. We assessed 25 full text papers, and excluded 17
this manner was included in the review. Evaluation of important (See Figure 1). Two studies (three reports) met our inclusion criteria
numerical data such as screened, randomised patients as well (Kirchhoff 2010; Perry 2005) Two studies reported both peritoneal
as intention-to-treat, as-treated and per-protocol population was and haemodialysis patients, and data could not be split from the

Advance care planning for haemodialysis patients (Review) 5

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

reports (Song 2010; SPIRIT Study 2013). Our attempts to contact the authors were unsuccessful; both studies have been added to
Studies awaiting classification.
 
Figure 1.   Study flow diagram

 
Included studies received usual care consisting of advance care planning facilitation;
standard advance directive counselling; assessment of advance
We included two studies (three reports) that involved a total of
directive on admission; and if patients needed more information.
337 patients and 134 surrogates (see Characteristics of included
Post-death data were reviewed to measure concordance between
studies).
patients' preferences and care at end of life.
Kirchhoff 2010 conducted a parallel RCT involving 313 patients and
Study outcomes included knowledge about advance care planning;
their surrogate pairs from 1 January 2004 to 31 July 2007. Of the 313
statement of treatment preferences; quality of patient-clinician
patients, 134 had diagnoses of ESKD and 179 were diagnosed with
communication about end-of-life care; and concordance of patient
end-stage coronary heart failure (CHF). Only those with ESKD were
preferences and care at end of life.
included in our analysis. Of the ESKD patients 70 were randomised
to the intervention group to receive patient centred-advanced care This study was published in two separate reports.
planning (PC-ACP) and 64 received usual care in the control group.
The study was conducted at two centres (six outpatient clinics of Perry 2005 conducted a three-arm parallel RCT that involved
large community or university health systems) in Wisconsin, USA. 203 participants, of whom 63 were assigned peer-mentoring
Participants were recruited from nephrology units, dialysis clinics intervention group, 59 to printed material intervention group, and
and dialysis units affiliated with the two study centres. 81 to the control group to receive routine care. The study was
conducted at 21 dialysis units in Michigan, USA.
Intervention group participants completed 60 to 90 minute
interviews with a trained facilitator. PC-ACP assessed patient and Study duration was two to four months for each participant.
surrogate understanding and experiences with illness; provided Participants randomised to the peer mentoring intervention group
information about disease-specific treatment options and their were contacted by peers eight times (five telephone calls and
benefits and burdens; assisted in documenting patient preferences three face-to-face meetings) to discuss the value of completing
for treatment; and assisted surrogate partners to make decisions advance directives. Peers had attended regional advanced directive
in line with patients' preferences. Control group participants workshops where advance directives issues were discussed. Peer

Advance care planning for haemodialysis patients (Review) 6

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

mentors and patients discussed the program; gave their own Excluded studies
experience with chronic illness; goals outside ESKD; spiritual
• Ten studies that were not RCTs (Badzek 2000; Choudry 1995;
orientation and fears; end-of-life considerations and barriers to
Fassett 2011; Friedman 2010; Hodge 2010; Katsube 2011;
completing advance directives; contribution to others and patient's
Lindblad 2010; Manda 2013; Parker 2009; Tangri 2011)
strength. Participants in the printed material received literature
developed by the (US) National Kidney Foundation. Control group • Four studies that compared interventions not relevant to this
participants received routine care provided by the dialysis unit (not review (den Hoedt 2013; Singer 1995; SPIRIT Pilot Study 2009;
described). Wright 2009)
• Two studies enrolled participants who were not relevant to this
Study outcomes included completion of advanced directive during review (Briggs 2004; Dallas 2012).
the study; whether or not patients wished to complete an advanced
directive if they had not done so during the study period; and Risk of bias in included studies
how comfortable the patient appeared to be in discussion about
We assessed risk of bias for both included studies. Overall, risk of
advance directives.
bias was assessed as unclear. (See Figure 2; Figure 3)
 
Figure 2.   Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages
across all included studies

 
 

Advance care planning for haemodialysis patients (Review) 7

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Figure 3.   Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Advance care planning for haemodialysis patients (Review) 8

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Allocation knew the degree of latitude (decision making authority a patient

wished to grant the chosen surrogate) the patient chose compared
Kirchhoff 2010 reported that pairs of eligible patients and their
with control group participants. Communication quality was rated
surrogates were randomised using the sealed-envelope method
as highly satisfactory among those in the intervention group,
and was assessed at low risk of bias. Perry 2005 did not report
and it was suggested that PC-ACP did not create anxiety or
randomisation or allocation concealment methods. Participants
other adverse effects. Kirchhoff 2010 also found higher rates of
were moved between groups after allocation and these domains
concordance between patients' preferences and end-of-life care
were judged to be at high risk of bias.
among intervention group participants compared with the control
Blinding group, including cardiopulmonary arrest.

Participants were not blinded in either study and were judged at Perry 2005 reported that peer support intervention resulted in a
high risk of performance bias. Neither Kirchhoff 2010 nor Perry 2005 higher proportion of participants completing an advance directive
report if outcome assessors were blinded and were assessed at or expressing a desire to complete, compared with printed material
unclear risk of bias for this domain. and routine care. The study also reported that peer support
intervention group participants had significantly greater levels
Incomplete outcome data of comfort about discussion of advance directives than printed
Both studies reported reasons and numbers of dropouts. Kirchhoff material and routine care group participants.
2010 conducted intention-to-treat analysis which was assessed at
PC-ACP and peer support may be useful interventions in advance
low risk of bias. Perry 2005 excluded participants without follow-up
care planning for haemodialysis patients. Both studies suggest that
data from analysis and was assessed at high risk of attrition bias.
in-depth discussion about advance directives and end-of-life care
Selective reporting did not cause unnecessary discomfort or anxiety. It should also
be noted that patients in Perry 2005 were younger than those in
No reporting bias was noted for either of the included studies and Kirchhoff 2010. However, the small number of included studies
all specified outcomes were reported. Protocols were unavailable. and participants means that further research is needed before
Reporting bias was assessed as low risk. conclusive evaluation can be made about advance care planning
for haemodialysis patients.
Other potential sources of bias
Kirchhoff 2010 reported four instances where neither surrogate Overall completeness and applicability of evidence
nor medical record information was available for both ESKA and Findings reported by Kirchhoff 2010 and Perry 2005 were confined
CHF participants. Kirchhoff 2010 did not reach calculated power to specific geographical and ethnic populations. Participants
levels due to resource limitations and the study was stopped were generally older people, and findings may therefore not be
early. Kirchhoff 2010 was supported by grants from the Agency of generalisable to diverse populations and settings where advance
Health Care Research and Quality and National Institutes of Health. care planning methods, healthcare systems and adherence rates
Two authors of Kirchhoff 2010 were employed by the Gundersen vary or do not exist. The small number of included studies meant
Lutheran Medical Foundation which owns rights to the intervention that we were unable to analyse data related to our specified
used in the study. We assessed Kirchhoff 2010 at high risk of other primary and secondary outcomes. We were unable to perform
bias. subgroup analysis or meta-analysis due to insufficient data.
Perry 2005 was supported by the Robert Wood Johnson Quality of the evidence
Foundation; National Kidney Foundation of Michigan; and
National Institute of Mental Health Career grant. Specific support Risk of bias was assessed as unclear or high in over half of
relationships were unclear. We assessed this study at unclear risk the domains assessed, limiting our confidence in the quality of
of other bias. evidence. Due to the nature of participants and interventions,
blinding of participants and interventionists was not possible and
Effects of interventions blinding of outcome assessment was not reported in either study.
Patients with missing follow-up data were excluded from the
Neither of the included studies reported on any of our specified
analysis by Perry 2005.
primary or secondary outcomes.
Due to insufficient resources, Kirchhoff 2010 was unable to recruit
DISCUSSION sufficient numbers of participants to attain optimal power during
the study period. Calculated power analysis required 560 patient-
Summary of main results surrogate pairs for optimal analysis; however, recruitment was
We included two studies (three reports) that involved a total of stopped at 313 pairs. The 179 participants with CHF were omitted
337 patients and 134 surrogates that investigated PC-ACP and peer from assessment for this review.
support interventions. Neither study reported any of our primary or
secondary outcomes. Overall, quality of evidence was assessed as suboptimal.

Kirchhoff 2010 investigated PC-ACP and reported that there was Potential biases in the review process
improved understanding of patient goals and preferences for Potential sources of biases in the review process include that
future medical decisions among surrogates who received PC- only a small number of studies were eligible for our review.
ACP compared with those in the control group. There was a Although literature searches included all major databases and
higher percentage of surrogates in the intervention group who
Advance care planning for haemodialysis patients (Review) 9
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

handsearching of kidney conference proceedings, it is possible that unnecessary distress and would help to enhance patients' quality
relevant unpublished data were not identified. of life at end-of-life. However, we found that the concept of advance
care planning was inconsistently defined in the literature, and
Agreements and disagreements with other studies or the lack published RCTs may be interpreted as an indication of
reviews suboptimal implementation of advance care planning in healthcare
settings. The effectiveness and value of advance care directives for
Our findings are consistent with an earlier audit of dialysis patients
people living with ESKD therefore remains largely unresolved.
by McAdoo 2012 where most deaths occurred in hospital settings
and dialysis was withdrawn before death among fewer than half of Implications for research
the participants.
Given the lack of evidence for advance care planning for people
The results of this review suggested that patients were highly receiving haemodialysis, large scale, well designed RCTs involving
satisfied with the quality of communication and greater levels of people with ESKD are necessary to determine the efficacy and
comfort. Our review is consistent with other studies suggesting that value of advance care planning for patients. Consistent methods for
discussion regarding advance care planning and end-of-life care reporting outcome measures that are patient-centred and relevant
did not destroy hope, cause unnecessary discomfort or anxiety to to health services should be further explored before widespread
patients (Davison 2006; Wright 2008). use can be justified. Additionally, consideration should be made
for development of a global standardised protocol for advance care
AUTHORS' CONCLUSIONS planning to facilitate its systematic evaluation.
Implications for practice ACKNOWLEDGEMENTS
The chronic and debilitating nature of ESKD indicates that advance
We would like to thank the referees for their feedback and advice
care planning should be considered for patients who choose to
during the preparation of this review. We would also like to thank
receive haemodialysis to prepare for future uncertainties while
Miss Jia Lin Chua, Miss Suyi Siow and Miss Khai Ho for their initial
cognitively unimpaired. Advance care planning does not cause
input to this review.

Advance care planning for haemodialysis patients (Review) 10

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

REFERENCES
 
References to studies included in this review Friedman 2010 {published data only}
Kirchhoff 2010 {published data only (unpublished sought but not Friedman EA. Stressful ethical issues in uremia therapy. Kidney
used)} International - Supplement 2010;78(117):S22-32. [MEDLINE:
20671740]
Kirchhoff KT, Hammes BJ, Kehl KA, Briggs LA, Brown RL. Effect
of a disease-specific advance care planning intervention on Hodge 2010 {published data only}
end-of-life care. Journal of the American Geriatrics Society
Hodge MH. The path to a paradigm shift in hemodialysis.
2012;60(5):946-50. [MEDLINE: 22458336]
Hemodialysis International 2010;14(1):5-10. [MEDLINE:
*  Kirchhoff KT, Hammes BJ, Kehl KA, Briggs LA, Brown RL. 20377648]
Effect of a disease-specific planning intervention on surrogate
Katsube 2011 {published data only}
understanding of patient goals for future medical treatment.
Journal of the American Geriatrics Society 2010;58(7):1233-40. Katsube M, Akisue Y, Nakamoto H. Recognition of Japanese
[MEDLINE: 20649686] nurse about “terminal care” and “right to die” in patients on
dialysis [abstract]. Peritoneal Dialysis International 2011;31(1
Perry 2005 {published data only} Suppl 1):S22. [EMBASE: 70716134]
Perry E, Swartz J, Brown S, Smith D, Kelly G, Swartz R. Peer
Lindblad 2010 {published data only}
mentoring: a culturally sensitive approach to end-of-life
planning for long-term dialysis patients. American Journal of Lindblad A, Juth N, Furst CJ, Lynoe N. When enough is enough;
Kidney Diseases 2005;46(1):111-9. [MEDLINE: 15983964] terminating life-sustaining treatment at the patient’s request: a
survey of attitudes among Swedish physicians and the general
  public. Journal of Medical Ethics 2010;36(5):284-9. [MEDLINE:
References to studies excluded from this review 20439330]
Badzek 2000 {published data only} Manda 2013 {published data only}
Badzek LA, Cline HS, Moss AH, Hines SC. Inappropriate use Manda K, Albert D, Germain M, Puntin G, Hakanson-Stacy J,
of dialysis for some elderly patients: nephrology nurses' Cohen L. Do ESRD patients really want to know prognosis?
perceptions and concerns. Nephrology Nursing Journal: [abstract]. American Journal of Kidney Diseases 2013;61(4):A62.
Journal of the American Nephrology Nurses' Association [EMBASE: 71024010]
2000;27(5):462-70. [MEDLINE: 16649321]
Parker 2009 {published data only}
Briggs 2004 {published data only}
Parker TF 3rd, Glassock RJ, Steinman TI. Conclusions,
Briggs LA, Kirchhoff KT, Hammes BJ, Song MK, Colvin ER. consensus, and directions for the future. Clinics of Journal of
Patient-centered advance care planning in special patient American Society of Nephrology: CJASN 2009;4 Suppl 1:S129-44.
populations: a pilot study. Journal of Professional Nursing [MEDLINE: 19995999]
2004;20(1):47-58. [MEDLINE: 15011193]
Singer 1995 {published data only}
Choudry 1995 {published data only}
Singer PA, Thiel EC, Naylor CD, Richardson RM, Llewellyn-
Choudry NK, Naylor CD. Reflections on supply-demand Thomas H, Goldstein M, et al. Life-sustaining treatment
mismatch in dialysis services in Ontario. CMAJ Canadian Medical preferences of hemodialysis patients: implications for advance
Association Journal 1995;153(5):575-81. [MEDLINE: 7641156] directives. Journal of the American Society of Nephrology
1995;6(5):1410-7. [MEDLINE: 8589316]
Dallas 2012 {published data only}
Dallas RH, Wilkins ML, Wang J, Garcia A, Lyon ME. Longitudinal SPIRIT Pilot Study 2009 {published data only}
Pediatric Palliative Care: Quality of Life & Spiritual Struggle Shields AM, Park M, Ward S, Song MK. Subject recruitment and
(FACE): design and methods. Contemporary Clinical Trials retention against quadruple challenges in an intervention trial
2012;33(5):1033-43. [MEDLINE: 22664645] of end-of-life communication. Journal of Hospice & Palliative
Nursing 2010;12(5):312-8. [PUBMED: 2010825388]
den Hoedt 2013 {published data only}
den Hoedt CH, Bots ML, Grooteman MP, Mazairac AH, Penne EL, Song MK, Ward S, Lin FC. End-of-life decision-making
van der Weerd NC, et al. Should we still focus that much on confidence in surrogates of African-American dialysis
cardiovascular mortality in end stage renal disease patients? patients is overly optimistic. Journal of Palliative Medicine
The CONvective TRAnsport STudy. PLoS ONE 2013;8(4):e61155. 2012;15(4):412-7. [MEDLINE: 22468770]
[MEDLINE: 23620729]
Song MK, Ward SE, Happ MB, Piraino B, Donovan HS,
Fassett 2011 {published data only} Shields AM, et al. Randomized controlled trial of SPIRIT: an
Fassett RG, Robertson IK, Mace R, Youl L, Challenor S, Bull R. effective approach to preparing African-American dialysis
Palliative care in end-stage kidney disease. Nephrology patients and families for end of life. Research in Nursing & Health
2011;16(1):4-12. [MEDLINE: 21175971] 2009;32(3):260-72. [MEDLINE: 19205027]

Advance care planning for haemodialysis patients (Review) 11

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Tangri 2011 {published data only} Davison 2006

Tangri N, Tighiouart H, Meyer KB, Miskulin DC. Both patient Davison SN, Simpson C. Hope and advance care planning in
and facility contribute to achieving the centers for Medicare patients with end stage renal disease: qualitative interview
and Medicaid services' pay-for-performance target for study. BMJ 2006;333(7574):886. [MEDLINE: 16990294]
dialysis adequacy. Journal of American Society of Nephrology
2011;22(12):2296-302. [MEDLINE: 22025629] Davison 2007
Davison SN, Torgunrud C. The creation of an advance care
Wright 2009 {published data only} planning process for patients with ESRD. American Journal of
Wright S, Bogan A, Chokshi A. DaVita program advocates Kidney Diseases 2007;49(1):27-36. [MEDLINE: 17185143]
self-management of CKD. Nephrology News & Issues
2009;23(13):28-9. [MEDLINE: 20050436] Davison 2010
Davison SN. End-of-life care preferences and needs: perceptions
  of patients with chronic kidney disease. Clinical Journal of The
References to studies awaiting assessment American Society of Nephrology: CJASN 2010;5(2):195–204.
Song 2010 {published data only} [MEDLINE: 20089488]
Song MK, Donovan HS, Piraino BM, Choi J, Bernardini J, Higgins 2003
Verosky D, et al. Effects of an intervention to improve
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring
communication about end-of-life care among African Americans
inconsistency in meta-analyses. BMJ 2003;327(7414):557-60.
with chronic kidney disease. Applied Nursing Research
[MEDLINE: 12958120]
2010;23(2):65-72. [MEDLINE: 20420992]
Higgins 2011
SPIRIT Study 2013 {published data only}
Higgins JP, Green S (editors). Cochrane Handbook for
Song MK, Lin FC, Gilet CA, Arnold RM, Bridgman JC, Ward SE.
Systematic Reviews of Interventions Version 5.1.0 [updated
Patient perspectives on informed decision-making surrounding
March 2011]. The Cochrane Collaboration, 2011. Available from
dialysis initiation. Nephrology Dialysis Transplantation
www.cochrane-handbook.org.
2013;28(11):2815-23. [MEDLINE: 23901048]
Holley 2003
Song MK, Ward SE. The extent of informed decision-making
about starting dialysis: does patients’ age matter?. Nephrology Holley JL. Advance care planning in elderly chronic dialysis
2014;27(5):571-6. [MEDLINE: 24519843] patients. International Urology & Nephrology 2003;35(4):565-8.
[MEDLINE: 15198168]
Song MK, Ward SE, Fine JP, Hanson LC, Lin FC, Hladik GA, et
al. Advance care planning and end-of-life decision making KDIGO 2013
in dialysis: a randomized controlled trial targeting patients Kidney Disease Improving Global Outcomes (KDIGO) CKD
and their surrogates. American Journal of Kidney Diseases Work Group. KDIGO 2012 clinical practice guideline for the
2015;66(5):813-22. [MEDLINE: 26141307] evaluation and management of chronic kidney disease.
Kidney International - Supplement 2013;3(1):1-150. [EMBASE:
Song MK, Ward SE, Lin FC, Hamilton JB, Hanson LC, Hladik GA, 2014145464]
et al. Racial differences in outcomes of an advance care
planning intervention for dialysis patients and their surrogates. Levey 2012
Journal of Palliative Medicine 2016;19(2):134-42. [MEDLINE: Levey AS, Coresh J. Chronic kidney disease. Lancet
26840848] 2012;379(9811):165-80. [MEDLINE: 21840587]
  Mallick 1999
Additional references
Mallick NP, Gokal R. Haemodialysis. Lancet
ANZDATA 2012 1999;353(9154):737-42. [MEDLINE: 10073530]
McDonald S, Clayton P, Hurst K. Thirty fifth annual report. 2012.
McAdoo 2012
www.anzdata.org.au/v1/report_2012.html (accessed 14 April
2016). McAdoo SP, Brown EA, Chesser AM, Farrington K, Salisbury EM,
pan-Thames renal audit group. Measuring the quality of end
Arulkumaran 2012 of life management in patients with advanced kidney disease:
Arulkumaran N, Szawarski P, Philips BJ. End-of-life care in results from the pan-Thames renal audit group. Nephrology
patients with end-stage renal disease. Nephrology Dialysis Dialysis Transplantation 2012;27(4):1548–54. [MEDLINE:
Transplantation 2012;27(3):879–81. [MEDLINE: 22379182] 21980155]

Brown 2007 Russon 2010

Brown E, Chambers EJ, Eggeling C. End of life care in Russon L, Mooney A. Palliative and end-of-life care in advanced
nephrology: from advanced disease to bereavement. Oxford: renal failure. Clinical Medicine 2010;10(3):279-81. [MEDLINE:
Oxford University Press, 2007. [ISBN/ISSN: 9780199211050] 20726463]

Advance care planning for haemodialysis patients (Review) 12

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Simon 2008 Wright 2008

Simon J, Murray A, Raffin S. Facilitated advance care planning: Wright A, Zhang B, Ray A, Mack JW, Trice E, Balboni T, et al.
what is the patient experience?. Journal of Palliative Care Associations between end-of-life discussions, patient mental
2008;24(4):256-64. [MEDLINE: 19227017] health, medical care near death, and caregiver bereavement
adjustment. JAMA 2008;300(14):1665-73. [MEDLINE: 18840840]
Vilar 2011
Vilar E, Farrington K. Haemodialysis. Medicine  
2011;39(7):429-33. [EMBASE: 2011354231] References to other published versions of this review
Lim 2013
Weisbord 2003
Lim CE, Siow S, Ho KE, Chua JL, Cheng NC, Kwok C, et
Weisbord SD, Carmody SS, Bruns FJ, Rotondi AJ, Cohen LM,
al. Advance care planning for haemodialysis patients.
Zeidel ML, et al. Symptom burden, quality of life, advance care
Cochrane Database of Systematic Reviews 2013, Issue 9. [DOI:
planning and the potential value of palliative care in severely
10.1002/14651858.CD010737]
ill haemodialysis patients. Nephrology Dialysis Transplantation
2003;18(7):1345–52. [MEDLINE: 12808172]  
* Indicates the major publication for the study
 
CHARACTERISTICS OF STUDIES

Characteristics of included studies [ordered by study ID]

 
Kirchhoff 2010 
Methods • Study design: parallel RCT·
• Study duration: 1 January 2004 to 31 July 2007
• Study follow up period: not stated

Participants • Country: USA

• Setting: National, 6 outpatient clinics of large community or university health systems associated with
two medical centres
• Relevant Health status:
* CHF, with New York Heart Association classification of II, III or IV
* ESKD, serum albumin concentration < 3.7 g/dL and a serious comorbidity such as diabetes melli-
tus, CHF, chronic obstructive pulmonary disease, a history of acute MI, or an above-the-knee am-
putation
• Total number: treatment group (160): control group (153)
* ESKD patients: treatment group (70): control group (64)
* CHF patients: treatment group (90): control group (89)
• Age: (mean ± SD)
* Patient: treatment group (71.4 ± 12.2): control group (70.6 ± 11.6)
* Surrogate: treatment group (59.5 ± 13.9): control group (57.4 ± 13.6)
• Sex: (M/F)
* Patient: treatment group (96/64): control group (89/64)
* Surrogate: treatment group (43/117): control group (41/112)
• Other relevant information: patients and their surrogate healthcare decision-makers were recruited
as pairs
• Exclusion criteria: < 18 years; did not have decision making capacity (had an activated power of attor-
ney for healthcare or determined through consultation with clinicians providing care to the patient);
unable to speak and understand English

Interventions Treatment group

• Patient Centred- Advance Care Planning (PC-ACP)

* PC-ACP interview delivered by trained facilitator including patient and surrogate lasting 60 to 90
minutes
* Resulted in documentation of patient goals for care in the medical record using a disease specific
Statement of Treatment Preferences (STP) form

Advance care planning for haemodialysis patients (Review) 13

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Kirchhoff 2010  (Continued)
Control group

• Usual care
* ACP facilitation with patient but no in-depth patient centred intervention in the presences of the
surrogate decision maker
* Standard advance directive counselling
* Assessment of advance directive on admission to the organisation
* Questions as to whether they would like more information

Outcomes • Knowledge about the ACP

• Statement of treatment preferences
• Quality of patient-clinician communication about end-of-life care
• Concordance of patient preferences and care at end of life

Notes • Funding source

* One author supported by grant from Agency of Health Care Research and Quality
* One author supported by grant from the Clinical and Translational Science Award program of the
National Centre for Research Resources, National Institutes of Health Grant in the final year of the
study
* Two authors were employed Gundersen Lutheran Medical Foundation Inc, which owns the rights
to the Respecting Choices program, of which the intervention used in the study, Disease Specific
Advance Care Planning is part
• Authors were contacted for further information and data specific to ESKD patients. However due to
technical reasons authors were unable to provide any additional data, hence only published data was
used in this review

Risk of bias

Bias Authors' judgement Support for judgement

Random sequence genera- Low risk Pairs of eligible patients and their surrogates were randomised to the control
tion (selection bias) group or the experimental PC-ACP group using sealed-envelopes

Allocation concealment Low risk Sealed envelopes

(selection bias)

Blinding of participants High risk Due to the nature of the interview, participants and interventionists could not
and personnel (perfor- be blinded
mance bias)
All outcomes

Blinding of outcome as- Unclear risk Did not report if outcome assessors were blinded
sessment (detection bias)
All outcomes

Incomplete outcome data Low risk All reasons and numbers of dropouts were reported. Intention-to-treat analy-
(attrition bias) sis conducted
All outcomes

Selective reporting (re- Low risk All stated outcomes reported

porting bias)

Other bias High risk There were 4 cases where there was neither surrogate nor medical record in-
formation was available (out of both CHF and ESKD patients). Two reviewers
blinded to group determined what treatment the patient received. For 14 cas-
es, where reviewers could not determine the care received or discrepancy was
present, it was referred to an independent Palliative Care Practitioner and ex-

Advance care planning for haemodialysis patients (Review) 14

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Kirchhoff 2010  (Continued)
pert interdisciplinary panel if necessary. Data from these were used in final da-
ta analysis

Study did not reach calculated power levels due to resource limitations and
hence was stopped early

Study was supported by grants from the Agency of Health Care Research and
Quality and National Institutes of Health. Two authors were employed by the
Gundersen Lutheran Medical Foundation which owns the rights to the inter-
vention used in the study

 
 
Perry 2005 
Methods • Study design: 3-arm parallel RCT
• Study duration: not stated
• Study follow up period: 2 to 4 months post intervention

Participants • Country: USA

• Setting: 21 dialysis units in Michigan
• Relevant health status: receiving HD (ESKD stage not specified)
• Total number: peer mentoring intervention (63), printed material (59), control (81)
• Age (mean (range)):
* Peer mentoring intervention: 44 (20 to 83 years)
* Printed materials: 44 (23 to 74 years)
* Control: 45 (20 to 80 years)
• Sex (M/F):
* Peer mentoring intervention: 29/34
* Printed materials: 32/27
* Control: 44/37
• Exclusion criteria: not stated

Interventions Treatment group

• Intervention 1: peer mentoring

* Patients assigned to receive peer intervention were contacted by peers on eight occasions that
included five telephone contacts and three face-to-face meetings
* Peers involved had previously attended a regional advanced directive workshop where issues sur-
rounding advance directives were discussed
* Over the course of the intervention, peer mentors and patients discussed the program; gave their
own experience with chronic illness; goals outside ESKD; spiritual orientation and fears; end-of-life
considerations and barriers to completing an AD; contribution to others and patient's strength
• Intervention 2: printed material
* Given printed material by the National Kidney Foundation (“Advance Directives: A guide for Pa-
tients and Families”)
* Distributed approximately at the midpoint of the two to four month study period
• Control
* No additional other means other than routinely provided by dialysis unit (not described)

Outcomes • Completion of advanced directive during study

• Whether patient wanted to complete an advanced directive if they had not so during the study period
• How comfortable the patient appeared to be in discussion advance directives (not comfortable (1),
very comfortable (5))

Advance care planning for haemodialysis patients (Review) 15

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Perry 2005  (Continued)
Notes • Funding source (specific support relationship not clear)
* Robert Wood Johnson Foundation
* National Kidney Foundation of Michigan
* National Institute of Mental Health Career grant

Risk of bias

Bias Authors' judgement Support for judgement

Random sequence genera- High risk Did not report randomisation method other than "Patients were assigned by
tion (selection bias) random lots". In 5 instances, patients assigned to the peer-intervention group
were moved to a control group by the social worker because of dialysis-sched-
ule conflicts that prevented peer intervention. In addition, the peer mentor at
1 unit became ill early in the study and could not carry out the peer interven-
tion; therefore, patients at this unit were assigned to the other 2 groups.

Allocation concealment Unclear risk Not reported

(selection bias)

Blinding of participants High risk Participants and social workers not blinded
and personnel (perfor-
mance bias)
All outcomes

Blinding of outcome as- Unclear risk Did not report if outcome assessors were blinded
sessment (detection bias)
All outcomes

Incomplete outcome data High risk Reasons and numbers for participant drop outs were reported. Patients with-
(attrition bias) out follow-up data were excluded. "A small number of patients did not have
All outcomes follow-up data and are excluded"

Selective reporting (re- Low risk All outcomes were reported

porting bias)

Other bias Unclear risk Supported by the Robert Wood Johnson Foundation; National Kidney Founda-
tion of Michigan; and National Institute of Mental Health Career grant. Specific
support relationships were unclear

AD - advance directive; CHF - chronic heart failure; ESKD - end-stage kidney disease; HD - haemodialysis; M/F - ,male/female; MI - myocardial
infarction; RCT - randomised controlled trial
 
Characteristics of excluded studies [ordered by study ID]
 
Study Reason for exclusion

Badzek 2000 Not RCT; descriptive cross-sectional study

Briggs 2004 Mixed population, unable to separate data for patients with ESKD

Choudry 1995 Not RCT; review article

Dallas 2012 Wrong population group; children with AIDS

Advance care planning for haemodialysis patients (Review) 16

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Study Reason for exclusion

den Hoedt 2013 Examined distribution of death causes in particular cardiovascular deaths; did not report advance
care planning

Fassett 2011 Not RCT; review article

Friedman 2010 Not RCT; review article

Hodge 2010 Not RCT; opinion/review

Katsube 2011 Not RCT; questionnaire

Lindblad 2010 Not RCT; questionnaire/survey

Manda 2013 Not RCT; chart review and survey

Parker 2009 Not RCT; summary and conclusions from the End Stage Renal Disease (ESRD) Steering Committee
meeting, did not contain data that could be assessed

Singer 1995 Crossover RCT; assessing which of 3 forms patients would use

SPIRIT Pilot Study 2009 Different interventions; studying recruitment and retention rates for palliative care studies

Tangri 2011 Not RCT; epidemiological study

Wright 2009 Wrong intervention; investigated education in self-management of CKD

AIDS - acquired immune deficiency syndrome; CKD - chronic kidney disease; ESKD - end-stage kidney disease; RCT - randomised controlled
trial
 
Characteristics of studies awaiting assessment [ordered by study ID]
 
Song 2010 
Methods • Study design: pilot parallel RCT
• Study duration: not stated
• Study follow up period: one week post intervention

Participants • Country: USA

• Setting: One dialysis clinic in Oakland, USA
• Relevant health status: Stage 5 CKD
• Number total: Intervention group (11): control group (8)
• Mean age ± SD (reported data not separated into intervention and control group)
* Patient: 52.82 ± 15.50
* Surrogates: 46.00 ± 10.12
• Sex (M/F) (reported data not separated into intervention and control group)
* Patient: 10/9
* Surrogates: 15/4
• Exclusion criteria: not stated

Interventions Treatment group

• Patient-centred advance care planning (PC-ACP)

* One hour in-depth interview with patient-surrogate dyad
* Delivered by trained nurse interventionist who had completed 2.5 days of PC-ACP training

Advance care planning for haemodialysis patients (Review) 17

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

Song 2010  (Continued)
Control group

• Usual care
* Nurse of social worker at dialysis clinic provided written information on advance directives;
addressed individual questions about life sustaining treatment options
* Questions about medical condition and related end-of-life treatment options referred to pa-
tient’s physician

Outcomes • Patient-Surrogate congruence in treatment preference

• Patients level of difficulty in making choices
• Surrogate's level of comfort in decision making
• Psychospiritual well-being of patient and surrogate

Notes • Funding
* University of Pittsburgh Central Research Development Fund
• Study reported both PD and HD patients and data could not be split from the reports; our attempts
to contact the authors were unsuccessful

 
 
SPIRIT Study 2013 
Methods  

Participants  

Interventions  

Outcomes  

Notes  

CKD - chronic kidney disease; HD - haemodialysis; M/F - male/female; PD - peritoneal dialysis

 

 
APPENDICES

Appendix 1. Electronic search strategies

 
 
Database Search terms

CENTRAL 1. ("advance care" next (plan or plans or planning)):ti,ab,kw

2. (advance next (directive* or decision*)):ti,ab,kw
3. (living next will*):ti,ab,kw
4. "right to die":ti,ab,kw
5. ((patient or patients) near/5 (advocat* or advocacy)):ti,ab,kw
6. "power of attorney":ti,ab,kw
7. (("end of life" or EOL) near/5 (care or discuss* or decision* or plan or plans or planning or prefer-
ence*)):ti,ab,kw
8. "terminal care":ti,ab,kw
9. (treatment near/5 (refus* or withhold* or withdraw*)).tw.
10.{or #1-#9}
11."renal replacement therapy":ti,ab,kw

Advance care planning for haemodialysis patients (Review) 18

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

  (Continued)
12.dialysis:ti,ab,kw
13.h*emodialysis:ti,ab,kw
14.("endstage kidney" or "endstage renal" or "end stage kidney" or "end stage renal"):ti,ab,kw
15.(ESKD or ESKF or ESRD or ESRF):ti,ab,kw
16.{or #11-#15}
17.{and #10, #16}

MEDLINE 1. exp Advance Care Planning/

2. (advance care adj (plan or plans or planning)).tw.
3. (advance adj (directive* or decision*)).tw.
4. living will*.tw.
5. Right to Die/
6. right to die.tw.
7. Patient Advocacy/
8. ((patient or patients) adj5 (advocat* or advocacy)).tw.
9. power of attorney.tw.
10.((end of life or EOL) adj5 (care or discuss* or decision* or plan or plans or planning or prefer-
ence*)).tw.
11.Terminal Care/
12.terminal care.tw.
13.Treatment Refusal/
14.exp Withholding Treatment/
15.(treatment adj5 (refus* or withhold* or withdraw*)).tw.
16.or/1-15
17.Renal Replacement Therapy/
18.renal replacement therapy.tw.
19.exp Renal Dialysis/
20.dialysis.tw.
21.(hemodialysis or haemodialysis).tw.
22.(endstage kidney or endstage renal or end stage kidney or end stage renal).tw.
23.(ESKD or ESKF or ESRD or ESRF).tw.
24.or/17-23
25.and/16,24

EMBASE 1. Living Will/

2. living will*.tw.
3. (advance care adj (plan or plans or planning)).tw.
4. (advance adj (directive* or decision*)).tw.
5. Right to Die/
6. right to die.tw.
7. Patient Advocacy/
8. ((patient or patients) adj5 (advocat* or advocacy)).tw.
9. "Power of Attorney"/
10.power of attorney.tw.
11.Terminal care/
12.((end of life or EOL) adj5 (care or discuss* or decision* or plan or plans or planning or prefer-
ence*)).tw.
13.terminal care.tw.
14.Treatment Refusal/
15.Treatment Withdrawal/
16.(treatment adj2 (refus* or withhold* or withdraw*)).tw.
17.or/1-16

Advance care planning for haemodialysis patients (Review) 19

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

  (Continued)
18.exp Renal Replacement Therapy/
19.renal replacement therapy.tw.
20.dialysis.tw.
21.(hemodialysis or haemodialysis).tw.
22.(endstage kidney or endstage renal or end stage kidney or end stage renal).tw.
23.(ESKD or ESKF or ESRD or ESRF).tw.
24.or/18-23
25.and/17,24

CINAHL S43 S36 AND S42 

S42 S37 OR S38 OR S39 OR S40 OR S41 

S41 TI (trial) 

S40 TI (double blind) OR AB (double blind) 

S39 TI (randomi?e*) OR AB (randomi?e*) 

S38 AB (placebo) OR AB (randomly) 

S37 (MH "Clinical Trials+") 

S36 S22 AND S35 

S35 S23 OR S24 OR S25 OR S26 OR S27 OR S28 OR S29 OR S30 OR S31 OR S32 OR S33 OR S34 

S34 TI (ESRF) OR AB (ESRF) 

S33 TI (ESRD) OR AB (ESRD) 

S32 TI (ESKF) OR AB (ESKF) 

S31 TI (ESKD) OR AB (ESKD) 

S30 TI (end stage renal) OR AB (end stage renal) 

S29 TI (endstage renal) OR AB (endstage renal) 

S28 TI (end stage kidney) OR AB (end stage kidney) 

S27 TI (endstage kidney) OR AB (endstage kidney) 

S26 TI (h#emodialysis) OR AB (h#emodialysis) 

S25 TI (dialysis) or AB (dialysis) 

S24 TI (renal replacement therapy) OR AB (renal replacement therapy) 

S23 (MH "Renal Replacement Therapy+") OR (MH "Continuous Renal Replacement Therapy+") OR
(MH "Hemodialysis+") OR (MH "Hemofiltration+") OR (MH "Home Dialysis") OR (MH "Peritoneal
Dialysis+") 

S22 S1 OR S2 OR S3 OR S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14

OR S15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21 

S21 TI (treatment N5 withdraw*) OR AB (treatment N5 withdraw*) 

S20 TI (treatment N5 withhold*) OR AB (treatment N5 withhold*) 

S19 TI (treatment N5 refus*) OR AB (treatment N5 refus*) 

S18 TI (terminal care) OR AB (terminal care) 

S17 TI (end of life) OR AB (end of life) 

Advance care planning for haemodialysis patients (Review) 20
Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

  (Continued)
S16 TI (power of attorney) OR AB (power of attorney) 

S15 TI (patient* N5 advocat*) OR AB (patient* N5 advocat*) 

S14 TI (right to die) OR AB (right to die) 

S13 TI (living will*) OR AB (living will*) 

S12 AB (advance N1 decision*) 

S11 TI (advance N1 decision*) 

S10 AB (advance N1 directive*) 

S9 TI (advance N1 directive*) 

S8 AB (advance care N1 plan) 

S7 TI (advance care N1 plan*) 

S6 (MH "Terminal Care+") 

S5 (MH "Patient Advocacy") 

S4 (MH "Treatment Refusal") 

S3 (MH "Right to Die") 

S2 (MH "Advance Directives+") 

S1 (MH "Advance Care Planning") 

Social Work Abstracts 1 (dialysis or hemodialysis or haemodialysis).mp.

2 (advance adj3 (care or plan or planning or directive*)).mp.

3 (power of attorney or end of life).mp.

4 2 or 3

5 1 and 2

 
Appendix 2. Risk of bias assessment tool
 
 
Potential source of bias Assessment criteria

Random sequence genera- Low risk of bias: Random number table; computer random number generator; coin tossing; shuf-
tion fling cards or envelopes; throwing dice; drawing of lots; minimization (minimization may be imple-
mented without a random element, and this is considered to be equivalent to being random)
Selection bias (biased alloca-
tion to interventions) due to High risk of bias: Sequence generated by odd or even date of birth; date (or day) of admission; se-
inadequate generation of a quence generated by hospital or clinic record number; allocation by judgement of the clinician; by
randomised sequence preference of the participant; based on the results of a laboratory test or a series of tests; by avail-
ability of the intervention

Unclear: Insufficient information about the sequence generation process to permit judgement

Allocation concealment Low risk of bias: Randomisation method described that would not allow investigator/participant to
know or influence intervention group before eligible participant entered in the study (e.g. central
allocation, including telephone, web-based, and pharmacy-controlled, randomisation; sequential-

Advance care planning for haemodialysis patients (Review) 21

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

  (Continued)
Selection bias (biased alloca- ly numbered drug containers of identical appearance; sequentially numbered, opaque, sealed en-
tion to interventions) due to velopes)
inadequate concealment of al-
locations prior to assignment High risk of bias: Using an open random allocation schedule (e.g. a list of random numbers); as-
signment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or
non-opaque or not sequentially numbered); alternation or rotation; date of birth; case record num-
ber; any other explicitly unconcealed procedure

Unclear: Randomisation stated but no information on method used is available

Blinding of participants and Low risk of bias: No blinding or incomplete blinding, but the review authors judge that the outcome
personnel is not likely to be influenced by lack of blinding; blinding of participants and key study personnel
ensured, and unlikely that the blinding could have been broken
Performance bias due to
knowledge of the allocated High risk of bias: No blinding or incomplete blinding, and the outcome is likely to be influenced by
interventions by participants lack of blinding; blinding of key study participants and personnel attempted, but likely that the
and personnel during the blinding could have been broken, and the outcome is likely to be influenced by lack of blinding
study
Unclear: Insufficient information to permit judgement

Blinding of outcome assess- Low risk of bias: No blinding of outcome assessment, but the review authors judge that the out-
ment come measurement is not likely to be influenced by lack of blinding; blinding of outcome assess-
ment ensured, and unlikely that the blinding could have been broken
Detection bias due to knowl-
edge of the allocated interven- High risk of bias: No blinding of outcome assessment, and the outcome measurement is likely to be
tions by outcome assessors. influenced by lack of blinding; blinding of outcome assessment, but likely that the blinding could
have been broken, and the outcome measurement is likely to be influenced by lack of blinding

Unclear: Insufficient information to permit judgement

Incomplete outcome data Low risk of bias: No missing outcome data; reasons for missing outcome data unlikely to be relat-
ed to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome
Attrition bias due to amount, data balanced in numbers across intervention groups, with similar reasons for missing data across
nature or handling of incom- groups; for dichotomous outcome data, the proportion of missing outcomes compared with ob-
plete outcome data. served event risk not enough to have a clinically relevant impact on the intervention effect esti-
mate; for continuous outcome data, plausible effect size (difference in means or standardized dif-
ference in means) among missing outcomes not enough to have a clinically relevant impact on ob-
served effect size; missing data have been imputed using appropriate methods.

High risk of bias: Reason for missing outcome data likely to be related to true outcome, with either
imbalance in numbers or reasons for missing data across intervention groups; for dichotomous
outcome data, the proportion of missing outcomes compared with observed event risk enough to
induce clinically relevant bias in intervention effect estimate; for continuous outcome data, plausi-
ble effect size (difference in means or standardized difference in means) among missing outcomes
enough to induce clinically relevant bias in observed effect size; ‘as-treated’ analysis done with
substantial departure of the intervention received from that assigned at randomisation; potentially
inappropriate application of simple imputation.

Unclear: Insufficient information to permit judgement

Selective reporting Low risk of bias: The study protocol is available and all of the study’s pre-specified (primary and
secondary) outcomes that are of interest in the review have been reported in the pre-specified way;
Reporting bias due to selective the study protocol is not available but it is clear that the published reports include all expected out-
outcome reporting comes, including those that were pre-specified (convincing text of this nature may be uncommon)

High risk of bias: Not all of the study’s pre-specified primary outcomes have been reported; one or
more primary outcomes is reported using measurements, analysis methods or subsets of the data
(e.g. subscales) that were not pre-specified; one or more reported primary outcomes were not pre-
specified (unless clear justification for their reporting is provided, such as an unexpected adverse

Advance care planning for haemodialysis patients (Review) 22

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
 Cochrane Trusted evidence.
Informed decisions.
 
 
Library Better health. Cochrane Database of Systematic Reviews

  (Continued)
effect); one or more outcomes of interest in the review are reported incompletely so that they can-
not be entered in a meta-analysis; the study report fails to include results for a key outcome that
would be expected to have been reported for such a study

Unclear: Insufficient information to permit judgement

Other bias Low risk of bias: The study appears to be free of other sources of bias

Bias due to problems not cov- High risk of bias: Had a potential source of bias related to the specific study design used; stopped
ered elsewhere in the table early due to some data-dependent process (including a formal-stopping rule); had extreme base-
line imbalance; has been claimed to have been fraudulent; had some other problem

Unclear: Insufficient information to assess whether an important risk of bias exists; insufficient ra-
tionale or evidence that an identified problem will introduce bias

 
CONTRIBUTIONS OF AUTHORS
1. Draft the protocol: DL, RN, NC, CK, FB, MC
2. Study selection: RN, DL, NC
3. Extract data from studies: RN, DL
4. Enter data into RevMan: RN, DL
5. Carry out the analysis: DL, RN, NC
6. Interpret the analysis: DL, RN, NC
7. Draft the final review: DL, RN, NC, CK, FB, MC
8. Disagreement resolution: DL, NC
9. Update the review: DL, NC, RN

DECLARATIONS OF INTEREST
• Chi Eung Danforn Lim: none known
• Rachel WC Ng: none known
• Nga Chong Lisa Cheng: none known
• Maria Cigolini: none known
• Cannas Kwok: none known
• Frank Brennan: none known.

SOURCES OF SUPPORT

Internal sources
• Nil, Other.

External sources
• Nil, Other.

INDEX TERMS

Medical Subject Headings (MeSH)

*Advance Care Planning;  *Renal Dialysis;  Hospitalization  [statistics & numerical data];  Kidney Failure, Chronic  [*therapy];  Randomized
Controlled Trials as Topic;  Third-Party Consent  [statistics & numerical data]

MeSH check words

Humans

Advance care planning for haemodialysis patients (Review) 23

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.