PAlliative Care in chronic Kidney diSease: the PACKS study—quality of life, decision making, costs and impact on carers in people managed without dialysis

The research study is a Post-Doctoral Fellowship funded by the National Institute
of Health Research in the UK. It is a mixed method study and includes quantitative
and qualitative components 37]. In the quantitative component a longitudinal survey of QOL, symptoms, cognition,
frailty, performance, satisfaction with decision-making, health service use of patients
and associated costs, subjective burden and QOL of carers, loss of earnings and the
opportunity cost of providing informal care will be explored. In the qualitative component
the decision making process with patients and carers that precedes referral to conservative
kidney management will be explored with renal physicians and/or clinical nurse specialists
in relation to patient satisfaction in decision making. The longitudinal nature of
the study will be explained to participants. Some people may change their mind regarding
their selected treatment option and commence dialysis although this is unusual in
clinical practice. If this happens the patient and carer data will continue to be
collected in order to examine how QOL might increase or decrease with a switch of
modality.

The study is multicentre across ten sites in Northern Ireland, England and Scotland.
Each site offers conservative management as a treatment option to patients who have
reached ESKD. All patients who make the decision not to embark on dialysis will be
approached to take part in the study. Their main carer will be asked to ‘opt-in’ to
the study with the patient’s consent.

Study approvals

The study protocol has been approved by the Office for Research Ethics Committees
Northern Ireland (ORECNI) (REC reference 14/NI/0057; 14 May 2014) and research governance
approval has been granted by the five Health and Social Care Trusts in Northern Ireland
and three acute hospital trusts in England, UK with two others in progress in London,
England and Glasgow, Scotland. The study is registered with the International Standard
Randomised Controlled Trial (ISRCTN) (ISRCTN06857980) 38]. It is sponsored by Queens University Belfast and the Western Health and Social Care
Trust, and has been adopted by the Northern Ireland Clinical Research Network (Renal).
It is supported by the Northern Ireland Clinical Trials Unit, a UK Clinical Research
Collaboration registered clinical trials unit.

Study population

Patients

To be eligible for recruitment into the study patients must be aged 18 years or older;
have reached stage 5 chronic kidney disease with estimated glomerular filtration rate??15 mL/min/1.73m
²
as measured by the MDRD formula 20]; and have made a confirmed decision for conservative kidney management, i.e. management
without dialysis or other renal replacement therapy. Patients lacking capacity to
give consent to participate will be excluded. Capacity for consent to participate
will be assessed in collaboration with the supervising nephrologists. Non-English
speaking patients or those who do not adequately understand verbal or written information
will be excluded.

Carers

To be eligible for recruitment into the study participants must identify as the primary
carer for the patient who has made a confirmed decision for conservative kidney management.
They must be over 18 years and the patient must agree that the carer can be approached
to participate. Carers can only participate in the study if they ‘opt in’ to study
by making contact with the Research Nurse. Carers who lack capacity to give consent
to participate in the study will be excluded. Non-English speaking carers will also
be excluded.

Renal physicians/clinical nurse specialists

Renal physicians and clinical nurse specialists (CNS) recruited to the study must
have experience of managing clinical consultations with patients who have stage 5
chronic kidney disease and are opting for conservative kidney management. Those clinicians
and nurses without experience of managing consultations with patients who opt for
conservative kidney management and who are not nephrologists or renal nurses will
be excluded.

Recruitment

Patients

Recruitment will take place over 18 months and data collection will continue for another
12 months or until death. The initial treatment decision is made in clinic with a
renal physician or a CNS. Once the decision is made Research Nurses at each site will
liaise with renal physicians or CNS and use the inclusion/exclusion criteria to identify
potential participants. All patients across the ten study sites who have made a decision
not to initiate dialysis and meet the study inclusion criteria will be sent written
information about the study (see Fig. 1). They will be contacted to see if they are willing to take part in the study. If
willing to participate written informed consent will be obtained at their next clinic
visit (Baseline). This should take place within 3 months after the initial decision
not to embark on dialysis.

Fig. 1. CONSORT diagram showing the flow of participants through each stage of study

If patients wish they can complete questionnaires themselves, with the nurses or with
their carers. The way in which the data are captured (self-completion or help from
staff member or carer) will be noted on the front sheet of the questionnaires.

Carers

With the patient’s agreement, and if the patient is happy for the carer to participate
in the study, information will be sent to the patient to give to their carer. Carers
will be asked to contact the Research Nurse if they are interested in ‘opting-in’
to the study. If a carer is interested in participating they will be invited to attend
the patient’s next clinic appointment so that informed consent can be obtained and
baseline data collected (see Fig. 1). If the carer is unable to attend this clinic visit then the Research Nurse will
send a consent form and baseline data collection tools to their home. The carers will
be asked to return the consent form to the hospital. Information on how to complete
the tools will be given over the telephone. Carers can either complete the tools themselves
or with the assistance of a Research Nurse.

Renal physicians/clinical nurse specialists

Renal physicians and CNS from each of the ten study sites who fulfil the inclusion
criteria will be identified through the lead nephrologists or senior renal managers
for each study site. The renal physicians and CNS will be invited to take part in
a semi-structured qualitative interview, conducted to explore their experience of
counselling a patient who ultimately declines dialysis (see Fig. 1). They will be approached by the Chief Investigator of the study via email. Written
informed consent will be obtained prior to the interview taking place. Recruitment
will continue until data are saturated 39].

Sample size

The primary outcome is to measure the true mean QOL at 3 months using the EQ-5D-5L
Visual Analogue Scale (EQ-VAS). The EQ-VAS asks patients to place their health on
an interval scale from 0–100 where 0 is the worst state they can imagine and 100 is
the best. Using a standard deviation estimate of 18 from a similar study (Nephrol
Dial Transplant (2004) 19: 1594–1599),with 100 individuals in our sample we would
calculate a 95 % confidence interval for the true mean which would be of width 7.2
units, i.e. 95 % of the time we would estimate the true mean within plus or minus
3.6 units. With an estimated death/dropout rate at 3 months of 10 %, a total of 112
patients are required. For the exploratory qualitative interviews with nephrologists
and CNS, in order to achieve data saturation when no new patterns emerge, a sample
of approximately 15–20 will be recruited until data saturation reached 39].

User involvement

The proposal has been developed in collaboration with members of the Kidney Research
and Education Initiative (KREI), established in 2010 to broaden collaborative working
between renal research groups and renal education teams to ensure involvement with
patients, carers, academics, clinicians and researchers 40]. In addition, we collaborated with the Northern Ireland Kidney Patient Association
(a self-help group run by patients with ESKD and their carers) to help set the study
aims and objectives. All those involved in this research proposal have personal knowledge
and experience of kidney disease and offer differing perspectives to the research
team. Regular feedback meetings will be arranged on the progress of the research for
comment and scrutiny with a particular focus on ethics, documentation (for example
in questionnaires, and information leaflets) and methodology. This shared decision-making
will strengthen the partnership between patients, clinicians and researchers involved
in the research. Several of the questionnaires were reviewed by members of the Northern
Ireland Kidney Patient Association and they provided constructive feedback on feasibility
of the study to the research team. The data collection tools were then piloted with
5 patients and 5 carers aged over 70 years attending a renal clinic to test for their
clarity, ease of use and time burden. Minor changes were made to the tools to be used
to collect economic information from patients and carers. All patients, carers, renal
physicians and CNS will have the right to withdraw from the study at any time.

Data collection

Patients

At baseline the research nurse will collect patient demographics through a review
of clinical records at study entry (Table 1). Data collected will include age, gender, ethnicity (using UK Office for National
Statistics categories), marital status, disease severity (as measured by eGFR) and
primary renal disease; comorbidity using the Davies Comorbidity Score. Information
will also be obtained on smoking habits of the patient, their diabetes status and
on the presence or absence of vascular disease (ischaemic heart disease, angina, myocardial
infarction, percutaneous coronary intervention, coronary artery bypass graft, peripheral
vascular disease and cerebrovascular accidents). Biochemical data at time of entry
to the study will be recorded including serum creatinine, albumin, calcium, phosphate,
parathyroid hormone, alkaline phosphatase, liver function tests, haemoglobin level
and urine protein creatinine ratio. The research nurse will also be asked to document
a ‘Yes’ or ‘No’ answer to the question, ‘would you be surprised if the patient dies
within 6 months?’ They will also complete the 9-point Clinical Frailty Scale, 6 item
Cognitive Impairment Test (6CIT) and Palliative Performance Score (PPS). Detail from
patients’ hospital notes will be collected on who made the decision to accept conservative
management. The Decisional Conflict Scale (DCS) will be completed at baseline and
6 monthly (see Table 2).

Table 1. Schedule of Assessments

Table 2. Data collection tools for PACKS study

Carers

At baseline and 3-monthly for 12 months, carers will be asked to complete, the EQ-5D-5L
whilst the patient is included in study or 3 months after death of patient or study
end; the EQ-5D-5L for the patient by proxy and a questionnaire to measure the subjective
and objective burden of providing care, loss of earnings and the opportunity costs
of providing care. The DCS will be completed at baseline and 6 monthly (see Table 2).

Renal physicians/clinical nurse specialists

Exploratory semi-structured qualitative interviews with consultant nephrologists and
CNS will be undertaken by the Chief Investigator over the period of the study to explore
the decision making process that precedes referral to conservative kidney management
in relation to patient satisfaction in decision making. Questions can be viewed in
Table 3.

Table 3. Interview guide for qualitative interviews with renal physicians/clinical nurse specialists

Outcome measures

The primary outcome of the study is QOL of patients at 3 months measured using the
EQ-VAS.

The range of data collection tools can be viewed in Table 2

Secondary outcome measures

Patients

Changes in QOL, symptoms, anxiety and depression in patients using the Kidney Disease
QOL (KDQOL) tool, EQ-5D-5L (which includes the EQ-VAS) and Palliative Outcome Scale
– Symptoms (POS-S) Renal 3-monthly over 12 months

Changes in cognition and frailty status in patients using the 6 Item Cognitive Impairment
Test (6CIT) and the 9-point Clinical Frailty Scale 3-monthly over 12 months

Changes in Performance in patients using the Palliative Performance Scale (PPS) 3-monthly
over 12 months

Patient Satisfaction in Decision Making using the Decisional Conflict Scale (DCS)
at baseline

Measurement of health and social care costs of patients receiving conservative kidney
management using a Patient Service Use Log 3-monthly over 12 months

Calculation of the number (%) of patient deaths 3-monthly and time to death.

Carers

Carer observation of patient’s Satisfaction in Decision Making using the Decisional
Conflict Scale (DCS) at baseline

Changes in QOL of carers using the EQ-5D-5L and Carer QOL 3-monthly over 12 months

Carers assessment of patient QOL using the EQ-5D-5L by proxy 3-monthly over 12 months

Subjective and objective burden of providing informal care in carers, loss of earnings
and the opportunity costs of providing informal care using items from the iMTA Valuation
of Informal Care Questionnaire 3 monthly over the 12 months.

Renal physicians/clinical nurse specialists

Exploration of the decision making process that precedes referral to conservative
kidney management in relation to patient satisfaction with renal physicians/CNS via
exploratory qualitative interviews.

The range of data collection tools can be viewed in Table 2

Quantitative data analysis

Descriptive statistics will be used to summarise baseline demographics and the questionnaire
data from patients and carers.

Primary outcomes

The mean and 95 % CI for the EQ-VAS will be calculated. Additional exploratory analyses
will be used to compare the mean EQ-VAS at 3 months between categories of categorical
variables (such as gender, marital status comorbidity) using ANCOVA adjusting for
baseline scores. The association between EQ-VAS nd other continuous variables will
be investigated using simple linear regression with EQ-VAS at 3 months as the dependent
variable. Multiple linear regressions will also be used to investigate associations
with comorbidities, age, gender, marital status, severity of symptoms and depression
to allow adjustments for confounding. As the study is not powered to detect statistically
significant differences in EQ-5D-5L VAS all results will be interpreted with caution.
Statistical significance will be assumed for P values 0.05.

Secondary outcomes

Descriptive summaries of patient and carer questionnaire data at 3, 6, 9 and 12 months
will be tabulated and presented graphically where appropriate. The number (%) of deaths
at 3, 6, 9 and 12 months will be tabulated. Time to death data will be investigated
using Kaplan Meier curves. The log rank test statistic will be used to compare categorical
variables and Cox proportional hazards model for continuous variables where appropriate.

Qualitative data analysis

In the exploratory qualitative aspect of the study, coding of qualitative data (renal
physician/CNS interviews) will be assisted with the use of NVivo version 10, qualitative
software to organise, store and retrieve data 41]. Renal physician/CNS interviews will be recorded by the researcher using a digital
recorder and subsequently professionally transcribed by a transcription company previously
employed to do similar work. Electronic versions of diary transcripts will be saved
and imported into the software programme to enable computer-assisted coding analysis.
This iterative process will be guided by an approach described by de Wet and Erasmus
42]. Their approach draws on grounded theory techniques including first level coding
and pattern coding and the development of relationships in the data.

Economic evaluation

A partial economic evaluation in the form of a cost-outcome description of conservative
kidney management will be performed. Each patient’s health and social care resource
use and quality of life measurements over the 12 month study period will be collected
as stated under Data Collection. Unit costs will be applied to the quantity of resource
use for each patient and these will be obtained from national sources where possible.
Utilities for the calculation of quality adjusted life years (QALYs) will be obtained
using responses on the EQ-5D-5L.

Care-related costs for the carers will also be explored. Multiple regression methods
will be used to examine patient and carer factors which are potentially associated
with their costs and to adjust for potential confounders. Although the cost effectiveness
of conservative kidney management compared with dialysis cannot be established within
the current single arm study design, the estimation of the costs and outcomes of conservative
kidney management will allow comparisons to be made with similar estimates for dialysis
already in the literature.

Study quality, monitoring and ethical considerations

To ensure the research quality and data is of a high standard rigorous study conduct
and monitoring procedures will be undertaken to ensure any problems are identified
and managed. Regular Trial Management Group meetings with the Northern Ireland Clinical
Trials Unit have taken place and will continue and an Advisory Group of experts has
met on two occasions and will continue to meet throughout the course of the study.
Formal monitoring of study procedures will take place at least once per year over
the course of the study at each site and as required.

Some patients in the study will be close to death. All attempts will be made to ensure
the best interests of patients and carers are taken into account. Permission to gain
access to patients will first be sought from staff. There is the potential for patients
and carers to feel coerced into the study due to the status of the research nurses
who are assisting with recruitment and data collection. This could intimidate participants
into agreeing to participate in the study 43]. One way of attempting to shift some of the power differentials is to explain honestly
to patients and carers that their care will not be compromised if they do not agree
to be involved in the study and it is entirely their decision. It will be necessary
to look for signs of fatigue or discomfort and to curtail any assessments as necessary.
Consideration will also need to be given to issues such as the length of assessment
and time of day when it takes place. Information (written and oral) will need to take
into account cognitive abilities and participants’ preferred way of assimilating information
and may need to be adapted appropriately e.g. some patients may need to be spoken
to slowly or need certain words explained. One view of dying people is that they are
in need of protection and should not be approached to be involved in research 44]. An alternative view is that everyone, whether dying or not should make the choice
for themselves. Throughout the study patients will be fully informed regarding the
research process in order to make appropriate decisions and retain the right to withdraw
at any time without care being compromised.