Get documents about "Living and Dying Well:
Shared by: HC1207300654
-
Stats
- views:
- 0
- posted:
- 7/29/2012
- language:
- pages:
- 35
Document Sample
DRAFT DECEMBER 2009
LIVING AND DYING WELL
SHORT LIFE WORKING GROUP 3
Development of recommendations for assessment tools for patients with
palliative and end of life care needs
Chair: Professor John Welsh
1
DRAFT DECEMBER 2009
Contents
Page
1 Remit of Short Life Working Group 3 3
2 Overview of sub groups (including draft recommendations)
2.1 Assessment tools for early identification of patients who may need 4
supportive/palliative care
2.2 Assessment tools to identify patients with increasing palliative care 5
needs and a likely prognosis of 6 or less months to live.
2.3 Assessment tools for symptoms 6
Further detail
3 Assessment tools for early identification of patients who may need 7
supportive/palliative care
4 Assessment tools to identify patients with increasing palliative care 9
needs and a likely prognosis of 6 or less months to live.
5 Assessment tools for symptoms 18
Annexes
A Gold Standards Framework – Prognostic Indicator Guidance 23
B References 27
C Additional prognostication tables 28
D Assessment tools for symptoms 34
E SLWG membership 35
2
DRAFT DECEMBER 2009
1 Remit of Short Life Working Group 3
The group’s remit was to develop recommendations relating to Advance Care Planning
and Assessment tools. In order to complete this task the group members divided into sub
groups and focused on the following areas:
1. Assessment tools for early identification of patients who may need palliative
care
2. Assessment tools to identify patients with increasing palliative care needs and
a likely prognosis of 6 months or less to live.
3. Assessment tools for symptoms
4. Advance Care Planning
The sections within this report reflect the work the subgroups 1 to 3. A separate report is
available on Advance Care Planning.
3
DRAFT DECEMBER 2009
2 OVERVIEW OF SUBGROUPS
2.1 Assessment tools for early identification of patients who may need
supportive / palliative care
There are many challenges and barriers to the initiation of appropriate palliative and end
of life care for patients across Scotland. As a consequence a significant number of
patients are not recognised as having palliative care needs until the final stages of their
illness, sometimes in the last few days of life. Thus the potential for effective palliative
and end of life care is often missed. The identification of people who are approaching the
palliative care phase of their illness provides the opportunity to raise questions and
issues about the expected course of illness, to begin to refocus care more onto
supportive and palliative care and address concerns and fears about dying.
If it were possible to predict prognosis accurately when a patient is in their last year of life
then a major barrier to the initiation of palliative care would have been overcome. This
would benefit a significant number of people in Scotland. The group were therefore
tasked with reviewing the literature with regard to predicting an early identification.
Methodology
The group were therefore tasked with reviewing the literature with regard to early
identification of palliative care patients.
Recommendations
Healthcare professionals across Scotland should be made aware of the Gold
Standards Framework –Prognostic Indicator Guidance (GSF-PIG) See Annex
A as a tool for highlighting patients with increasing palliative care needs and
limited prognosis:
-The GSF_PIG should be used to help identify patients whose health
is declining and would potentially benefit from introducing the
palliative care approach, and,
-may be ‘at risk’ of dying in the next 12 months. (the surprise
question)
Patients identified by clinicians from any healthcare setting using the GSF-
PIG should be highlighted and placed on the General Practice Palliative Care
Register.
Patients identified by using the GSF-PIG should be assessed with regard to
their palliative care needs and advance care planning commenced.
4
DRAFT DECEMBER 2009
2.2 Assessment tools to identify patients with increasing palliative care needs
and a likely prognosis of 6 or less months to live.
Diagnosis, treatment and estimation of prognosis are skills expected of doctors and other
health care professionals. Palliative care has burgeoned in the past 30-40 years and
changes in health care policy have resulted in a greater demand and tendency to
address the length of survival of patients with incurable progressive conditions. In
addition to identified patients and carers receiving the required support from the statutory
and voluntary sectors there are many advantages for clinicians and patients / carers to
be aware of an estimate of survival.
Methodology
An extensive literature search was undertaken, Database: Ovid MEDLINE(R) <1996 to
January Week 2 2009> Other databases were searched and various articles from
retrieved papers further appraised.
Recommendations
The PPS v2 alone or in combination with the PPI should be adopted in Scotland
as the predictive tool for approximately 6 month or less survival.
The survival nomogram should be used to calculate median survival period and
estimation of range of survival
5
DRAFT DECEMBER 2009
2.3 Assessment tools for symptoms
The sub- group’s remit was to examine and summarise the evidence for the use of
assessment tools for symptoms in palliative care and make recommendations based on
this.
Palliative care patients experience a number of symptoms, which affects the overall QoL
of patients. Accurate and regular assessment of symptoms is important for early
identification and management of symptoms. Symptoms such as pain are better
assessed than others, and symptom assessment and management is more advanced in
certain patient groups, for e.g. cancer patients.
Methodology
Several assessment tools are in use and there is no clear consensus or guidelines
recommending any particular tools. The group’s aim was to review the literature on this
topic and make a recommendation.
The sub-group concentrated on assessment tools for pain and spiritual distress
Recommendations
There is no single symptom assessment scale that will fit all settings and
meets all the criteria for an ideal assessment tool.
The four tools below have been validated (Annex D)
The Edmonton Symptom Assessment Scale (ESAS)
M.D. Anderson Symptom Inventory (MDASI)
Condensed MSAS (CMSAS)
The Cambridge Palliative Assessment Schedule ( CAMPAS-R)
The choice of tool will depend on the trade off between various features and
characteristics which are considered essential to a particular setting
6
DRAFT DECEMBER 2009
3 Assessment tools for early identification of patients who may need
supportive / palliative care
To advise on appropriate assessment tools to guide early identification in order to assess
and initiate appropriate end of life care practice the subgroup undertook an extensive
Medline search for literature on 1 year prognosis, E-mail correspondence with colleagues
across Scotland and entered numerous round table discussion.
The literature search confirmed that identifying when a person has a 1 year prognosis is
difficult for those with cancer and significantly more so for those with organ failure and
dementia.
There are several tools for predicting prognosis at 12 months, however none are
particularly robust. One reason for their lack of robustness is that no single tool is
capable of accurately predicting prognosis across the 3 well-described disease
trajectories.
SPARRA (Scottish patients at risk of readmission and admission) data now exists in
Scotland highlighting those patients at high risk of readmission. Although this data is not
a prognostic tool, repeated hospital admission over a short time-scale is often a poor
prognostic indicator and it would be reasonable to assess those patients with a high
SPARRA score to consider whether they are predicted to be in the ‘at risk’ of dying
category and if their palliative care needs have been met.
The Gold Standards Framework, Prognostic Indicator Guidance (GSF-PIG), highlights
specific 12-month prognostic markers for several common disease processes. Individual
PIG charts are available for a range of conditions e.g. heart failure, cancer, Parkinson’s
disease.(Annex A) Several bodies have adopted this guidance across the United
Kingdom and it would seem appropriate to adopt use of this tool in Scotland. This group
sought the opinions of specialists in various disciplines with regard to their knowledge of
and the potential usefulness of the GSF-PIG. The consensus was that the GSF-PIG was
not as specific or sensitive for predicting 12 month prognosis as many would like.
However it was considered to be extremely useful in highlighting patients whose health
was declining and who would potentially benefit from adopting a palliative approach to
their care.
This working group considered that using the GSF-PIG to identify palliative patients was
extremely valuable, regardless of the accuracy of predicting 12 month prognosis. The
result of identifying these patients is that an assessment of need would be performed and
steps taken to appropriately meet their needs.
Recommendations
Healthcare professionals across Scotland should be made aware of the Gold
Standards Framework –Prognostic Indicator Guidance (GSF-PIG) See Annex
7
DRAFT DECEMBER 2009
A as a tool for highlighting patients with increasing palliative care needs and
limited prognosis:
-The GSF_PIG should be used to help identify patients whose health
is declining and would potentially benefit from introducing the
palliative care approach, and,
-may be ‘at risk’ of dying in the next 12 months. (the surprise
question)
Patients identified by clinicians from any healthcare setting using the GSF-
PIG should be highlighted and placed on the General Practice Palliative Care
Register.
Patients identified by using the GSF-PIG should be assessed with regard to
their palliative care needs and advance care planning commenced.
8
DRAFT DECEMBER 2009
4 Assessment tools to identify patients with increasing palliative care needs
and a likely prognosis of 6 or less months to live.
Diagnosis, treatment and estimation of prognosis are skills expected of doctors and other
health care professionals. Palliative care has burgeoned in the past 30-40 years and
changes in health care policy have resulted in a greater demand and tendency to
address the length of survival of patients with incurable progressive conditions. In
addition to identified patients and carers receiving the required support from the statutory
and voluntary sectors there are many advantages for clinicians and patients / carers to
be aware of an estimate of survival.(1)
An extensive literature search was undertaken, Database: Ovid MEDLINE(R) <1996 to
January Week 2 2009> Other databases were searched and various articles from
retrieved papers further appraised.
The sub group’s conclusion is that patient’s palliative care needs are not exclusively
linked to survival time and as a result:
The group unanimously felt that recognition of patients’ and carers’ needs was the
paramount factor
Careful appraisal of the obtained literature produced evidence that prediction of
prognosis was poor with the majority of studies showing that doctors overestimated
survival (2). The literature also highlighted:
Most studies on prognostication have been with cancer patients.
The rate of decline in a patient’s condition has a bearing upon survival time.
Probabilistic predictions (50-75% accuracy) are likely to be more accurate than
predicting remaining survival time (25% accuracy).
150 variables have been evaluated in the literature for their ability to predict survival. (3)
The variables examined in an endeavour to predict survival can be allocated to 3
parameters ;
1. Biological related
2. Performance related
3. Symptom related
1. Biological - The Glasgow Prognostic Score
McMillan has reviewed (4) the literature and published (5) research on the relationship of
degree of systemic inflammation, nutritional status and survival in cancer patients. The
Glasgow Prognostic Score based upon the acute phase protein, C-reactive protein (CRP)
and the patient’s albumin levels has been shown to be of value in estimating prognosis.
Using these two parameters in comparison to the commonly used performance status
9
DRAFT DECEMBER 2009
and stage of cancer has shown a strong statistically comparable accuracy in
prognostication (6)
2. Performance status, is the single most extensively researched parameter.
For example;
The Karnosky Performance Scale (KPS)
The Eastern Cooperative Oncology Group (ECOG) - Performance Status Scale
Both have been associated with prediction of survival demonstrating that the
poorer the score the shorter the prognosis.(7,8)
Palliative Performance Scale (PPS v2) (9)
This performance indicator has been produced and developed by the Victoria Hospice
Society, BC.. It is a more detailed tool based upon the KPS. Use of the PPS has shown a
reliable correlation between PPS score and survival. See Table 1
The PPS has 11 grades ranging from 0-100%. Patients with 0% score are dead and
those scoring 100% are currently well and independent. There is a high and reliable
probability that of those with a score of 50% or less only 10% will be alive in 6 months.
See Graph 1
Table 1 Showing relationship of PPS score to survival time
No. of Survival Time in Days
Initial PPS
Patients Median C.I. Mean Mode Range
PPS 100% - - - - - -
PPS 90% - - - - - -
PPS 80% - - - - - -
PPS 70% 129 63 48.4, 77.6 140.7
PPS 60% 453 43 37.9, 48.1 92.1
PPS 50% 1,003 28 25.1, 30.9 75.9
PPS 40% 1,590 13 11.9, 14.1 39.1
PPS 30% 1,403 5 4.6, 5.4 20.1
PPS 20% 732 2 1.8, 2.2 7.3
PPS 10% 569 1 0.9, 1.1 3.0
Overall 5,879 8 7.5, 8.5 40.3
10
DRAFT DECEMBER 2009
Graph 1 Showing relationship of PPS score to probable survival time
...
Details of the scoring system and definitions are shown below with the permission of the
Victoria Hospice Society
11
DRAFT DECEMBER 2009
12
DRAFT DECEMBER 2009
13
DRAFT DECEMBER 2009
The survival nomogram which takes other factors into consideration including diagnosis,
age and sex is illustrated in Fig 2 below
Fig. 2. A survival nomogram based on age, gender, location, diagnosis, and PPS.
To illustrate its use, consider the example of an 86-year-old male patient with initial PPS
of 20%, admitted from home PRT with noncancer cardiovascular disease. The covariate
points are: AGE = “e” for 10 points, GENDER = “M” for 8 points, LOCATION = “home
PRT” for 10 points, DX = “o” for 3 points, PPS = “20%” for 77 points, for total points of
108. Then, drawing a vertical line downward from the total points calculated yields a
median survival of just over two days, first (75th) quartile of just over four days, and third
(25th) quartile of one day
The above tables graphs and nomogram are taken for a paper by Lau et al.(9)
Lau et al (9) have recently published a review of prognostication using the PPS in 6066
patients admitted to a specialist palliative care service. 941 of these patients did not have
cancer. They found that the PPS predicted survival at a significant level (p>0.05).
Increased survival times were associated with higher PPS scores. Age, gender, location
and diagnosis were all significant predictors of survival time. This study reported KM
14
DRAFT DECEMBER 2009
survival time distributions and life expectancy based upon an aggregation of all ages,
gender, locations and diagnosis for the participants . However the survival time
nomogram is based on the Cox regression model adjusted for all the variables present.
So the survival time taken from the nomogram is independent of diagnosis. Hence the
applicability of the nomogram to both malignant and non-malignant conditions
A study was performed by Weng (10) using the Palliative Performance Scale (PPSv2) in
a retrospective chart audit of 492 cancer patients admitted over an 18-month period to a
minority-serving home-based hospice and palliative care program. Applying the PPSv2 to
a predominantly minority ethnic sample produced results that are similar to those from
predominantly Caucasian samples.
3. Symptom related - A further tool is the Palliative Prognostic Index (PPI). This
combines performance with symptoms of dyspnoea and delirium along with the presence
of oedema and assesses oral intake ability. Combining the PPSv2 and PPI yields a
predictive accuracy with an 80% sensitivity and an 85% specificity.
Recommendations
The PPS v2 alone or in combination with the PPI should be adopted in Scotland
as the predictive tool for approximately 6 month or less survival.
The survival nomogram should be used to calculate median survival period and
estimation of range of survival
Recommendations for use.
In patients with progressive and incurable conditions who present with or exhibit triggers
suggestive of palliative care needs such as repeated hospital admissions over a short
time scale, repeated house call requests, declining functional ability then among other
actions which should be considered an estimate of likely survival can be made if relevant,
using the PPS nomogram.
This simple and quick tool should be useable in all settings, namely
Acute hospitals
Hospices
Peoples’ homes
Care homes
Outcome of assessment tool use.
A sequence of events will be triggered to ensure full palliative care assessment occurs.
Events being triggered will embrace the Advance Care Planning philosophy and will
depend upon needs identified but may include any or all of the following:
15
DRAFT DECEMBER 2009
If palliative care need and/or poor prognosis is identified in the acute setting
the patient’s GP should be notified and placed on GSF register if not already
recognised
If palliative care need and/or poor prognosis is identified in community setting,
GP will place patient on GSF register
District Nurse assessment and input
Financial assessment and facilitation of receipt of full welfare entitlement
Social service input such as home help and/or Occupational Health
assessment and appropriate input
Physiotherapy assessment and appropriate input
Palliative medicine consultant involvement - acute setting- community setting
CNS palliative care involvement–acute setting- community setting
Hospice involvement
Pastoral care assessment and appropriate input
Marie Curie community nursing input
NHS equipment provision
Completion of an Anticipatory Care Plan, e-PCS, Thinking ahead patient
preference plan
Possible means of monitoring results of applying this tool.
Quality of life assessed by each individual patient (e.g. EORTC-QOL ) should
improve
Unnecessary and potentially futile treatment may be avoided
Patients and carers can discuss their future wishes at a stage reasonably
distant from imminent death.
If it is recognised that survival is relatively short, and this information sensitively
imparted , patients have the opportunity to exert more control over their
remaining life span.
Challenges
Gaining public acceptance/cultural shift to facilitate more open discussion
concerning one’s mortality
Education of NHS/Voluntary sector healthcare providers and social services
Incorporation of routine screening in all with palliative conditions
Notification of score and implication to all involved with patient’s care
Having resources to meet identified needs
Suggested resource implications
16
DRAFT DECEMBER 2009
Education and Training
Personnel to complete tool
Raising public awareness and expectations
A small pilot project examining this service development is underway. Much of the early
work is with cancer patients and the prognostic accuracy of the tool/tools needs to be
further studied prospectively in more cancer and non-cancer patients.
17
DRAFT DECEMBER 2009
5 Assessment tools for symptoms
The evidence base for this section of the report is based on evidence from a literature
review, communications with experts across Scotland and round table discussions. The
search strategy for the literature review was as follows;
Database: CDSR, ACP Journal Club, DARE, CCTR, CLCMR, CLHTA, CLEED,
Ovid MEDLINE(R), EMBASE, PsycINFO, Journals@Ovid
Search Strategy: Forty three articles were identified by the above search strategy. These
were extracted on to an excel spread sheet. Most of the research on symptom
assessment in palliative care was done on patients with cancer.
Abstracts included a recent systematic review on symptom assessment scales in cancer.
These were scrutinised for articles that reported on use of assessment tools in palliative
care. Scales referred to in these articles were noted down. All these scales were
examined against the following arbitrary criteria set by the subgroup;
validated in palliative care population
includes physical and psychological symptoms
practicality- brief and easy to complete
self and observer rating possible
Desirable features were;
Simple and can be completed visually
accommodate repeated measurements
used extensively in palliative care
Applying these criteria, we identified six potential symptom scales.
1. The Edmonton Symptom Assessment Scale (ESAS)
2. M.D. Anderson Symptom Inventory (MDASI)
3. Memorial Symptom Assessment Scale (MSAS)
4. Condensed MSAS (CMSAS)
5. The Palliative Care Outcome Scale (POS)
6. The Cambridge Palliative Assessment Schedule ( CAMPAS-R)
Another scale used locally at the cancer centre was also considered- The Psychological
Medicine Research Symptom Check Up
The findings suggested that symptoms can be assessed either as a single dimension
(e.g. frequency- “how often”) or multiple dimensions (i.e. frequency, duration, severity
and distress). How and which of these contribute to an individual’s QoL is not known.
18
DRAFT DECEMBER 2009
Symptoms are assessed over a variety of time scales from “now” to “past two weeks”. A
shorter time frame may be appropriate for assessment of symptoms, which are liable to
change. A shorter time scale may also be most appropriate for patients with cognitive
impairment. However flexible time scales that can be used in different settings such as
community or in patients was considered desirable.
A systematic review of symptom assessment tools in cancer 1 listed the characteristics of
an ideal symptom assessment tool below.
Characteristics of an ideal symptom assessment instrument (Kirkova et al)
However the review did not find any tool that fitted all the characteristics. This was felt to
be the case in our review as well. The six selected tools were further scrutinised against
some pragmatic criteria, further detail is available that compares the tools against these
criteria.
Special considerations: Symptom assessment in patients with cognitive impairment
and/or communication difficulties and assessment of spiritual distress emerged as
important considerations in the group discussions. The initial review was supplemented
with another review of literature and communications with experts across Scotland to
summarise evidence, practice and make recommendations.
Symptom assessment in patients with cognitive impairment: Self assessment scales
have been validated in older populations and remain the gold standard for the evaluation
of pain intensity in this age group. Most patients with dementia demonstrate appropriate
use of self-assessment scales. Observational scales correlate moderately with self-
assessment and tend to underestimate pain intensity; thus their use should be reserved
for patients who have demonstrated their inability to use self assessment tools reliably.
Pain Assessment for Older Adults:
Patients self report is the most reliable measure of pain intensity as there are no biology
markers of pain. Simply worded questions and tools, which can be easily understood, are
the most effective, as older adults frequently encounter numerous factors, including
sensory deficits and cognitive impairments.
19
DRAFT DECEMBER 2009
The most widely used pain intensity scales used with older adults are the Numeric Rating
Scale (NRS), the Verbal Descriptor Scale (VDS) and the Faces Pain Scale-revised (FPS-
R). The most popular tool, the NRS, asks the patient to rate their pain by assigning a
numerical value with zero indicating no pain and 10 representing the worst pain
imaginable. The VDS asks the patient to describe their pain from “no pain to pain as bad
as it could be”. The FPS-R asks patients to describe their pain according to a facial
expression that corresponds with their pain.
Validity and Reliability: Amongst these three scales, several studies have demonstrated
concurrent validity between 0.56 and 0.90 with the lowest correlations found between the
FPS-R and the other scales suggesting the FPS-R may be measuring a broader
construct incorporating pain.
Strengths and limitations: Overall the NRS was the preferred scale with cognitively intact
older adults and the FPS-R was the preferred scale with cognitively impaired patients.
These brief assessment tools should not replace performing a comprehensive health
history and physical exam, which may lead to the determination of etiologies of pain
Behavioural Pain Assessment Tools: Recent critiques of behavioural pain assessment
tools commonly used with older adults with end stage dementia indicate that there is
currently no tool that has sufficient reliability and validity to support broad adoption in
clinical practice, Published studies evaluating existing tools are largely based on limited
testing conducted in small samples or in a limited number of clinical settings. One
comprehensive review of tools may be accessed at www.cityofhope.org/prc/elderly.asp
and as an article by Herr, Bjoro and Decker (2006) Tools for assessment of pain in
nonverbal older adults with dementia: a state of the science review Journal of Pain
Symptom Management 31 `170-192
Existing Behavioural Pain Scales for Persons with Dementia
Abbey Pain Tool
*ADD Protocol (Assessment of Discomfort in Dementia)
CNPI (Checklist of Nonverbal pain Indicators)
Doloplus – 2 Scale
DS-DAT (the Discomfort Scale – Dementia of the Alzheimer’s Type)
FLACC (Face, Legs, Activity, Cry, Consolability)
NOPPAIN (The non-Communicative Patients pain Assessment Instrument
PACSLAC (The Pain Assessment Checklist for Seniors with Limited Ability to
Communicate)
PADE (Pain assessment for the Dementing Elderly)
PAINAD (Pain Assessment in Advanced Dementia)
RaPID (Rating Pain in Dementia)
* ADD is not an assessment tool, but a protocol for validating presence of pain
20
DRAFT DECEMBER 2009
Assessment of Spiritual distress: Assessment of spiritual need in health care is
relatively recent and is reflected by limited literature on the topic. Clinical spiritual care
uses various assessment strategies to elicit a person’s spiritual orientation, beliefs and
needs. Patients may bring this up as a concern, patient’s can be systematically asked to
complete a self report or clinicians can do a spiritual inquiry or assessment . 2 An opinion
review on the topic of spiritual assessment in health care recommends viewing spiritual
assessment as a process which is practically relevant and advices caution on the use of
stand alone assessment tools. 3
A summary of these approaches with some relevant tools are listed on the NCI PDQ
summary on the topic “Screening and Assessment of Spiritual Concerns” 1
Another summary approaches in mental health care is attached as a guide. Two
validated tools were identified from discussions with experts. These were the:
The Royal Free Questionnaire for Spiritual and religious beliefs and
Self Perception and Relationship Tools (S-PRT)
Recommendations for assessment tools for symptoms
There is no single symptom assessment scale that will fit all settings and meets all the
criteria for an ideal assessment tool. The four tools identified, listed below, that are
validated and can be used in a palliative care setting and available to view in Annex D.
The Edmonton Symptom Assessment Scale (ESAS)
M.D. Anderson Symptom Inventory (MDASI)
Condensed MSAS (CMSAS)
The Cambridge Palliative Assessment Schedule ( CAMPAS-R)
The four recommended tools that should be used- ESAS, CMSAS, MDASI and
CAMPAS-R. The choice would depend on the trade off between various features and
characteristics which are considered essential to a particular setting e.g. choosing a
scale with verbal rating scale for people with cognitive impairment.
Spiritual needs or distress assessment techniques vary from use of standardised tools to
semi structured interview techniques. Validated tools are available and the choice would
depend on the purpose of evaluation, personnel conducting the assessment and
resources available to manage any spiritual concerns.
Based on the current work we are unable to recommend any particular tool.
Symptom assessment in cognitively impaired patients is a challenge. As far as possible
they should be encouraged to complete self reports in simple formats such as the NRS or
VDS. Most of the recommended symptom assessment scales are in this format.
Observational or behavioural assessment scales can be used in people who fail to
complete self reports.
There is not enough evidence to recommend any of the behavioural tools at the moment
21
DRAFT DECEMBER 2009
All the recommended scales have items that screen for psychological distress in the form
of anxiety, sadness or depression. A positive response to these should be followed up by
a more detailed psychological/psychiatric assessment. Another approach is to have two
stage screening system, where a positive response to these single items is followed up a
more detailed self report such as Hospital Anxiety and Depression Scale ( HADS). A high
score on this then can be followed up by detailed psychological/psychiatric assessment.
Further considerations identified through the subgroup: A major challenge with self
report assessment scales is to ensure its regular and effective use. Simplicity, ease of
use, acceptability by patients and staff are important considerations. Electronic and semi
automated systems with live feedback are shown to be acceptable and feasible.
Another consideration with self reports is determining clinical significance to the
responses or scores. Determining the clinical significance and impact on daily living and
interaction with others.
Healthcare professionals need to be trained in the scoring and interpretation of scales
and in managing symptoms effectively. Electronic adaptation requires technological skills
and resources.
Further research should be considered in the following areas:
Symptom dimensions (frequency, severity, duration and distress) and their
relationship to symptom burden and QoL.
Symptom assessment in palliative patients with other chronic illnesses other than
cancer
Symptom assessment in patients with cognitive impairment and assessment of
spiritual needs and distress.
22
DRAFT DECEMBER 2009
Annex A
23
DRAFT DECEMBER 2009
24
DRAFT DECEMBER 2009
25
DRAFT DECEMBER 2009
26
DRAFT DECEMBER 2009
Annex B - References / Bibliography
1. Christakis NA, Lamont EB. Extent and determinants of error in doctors’ prognoses in
terminally ill patients: prospective cohort study. BMJ.2000;320(7233):469-72
2. Glare P,Virik K, Jones M. A systematic review of clinicians survival predictions in
terminally ill cancer patients BMJ 2004;327: 195-205
3. Vigano A, Dorgan M, Buckingham J, Bruera E, Suarez-Almazor ME. Survival prediction
in terminal cancer patients: a systematic review of the medical literature. Palliat Med
2000;14(5):363-74)
4. McMillan DC. Systemic inflammation, nutritional status and survival in patients with
cancer. Current Opinion in Clinical Nutrition and Metabolic Care.2009;12:223-226
5. McMillan DC. An inflammation-based prognostic score and its role in the nutrition-based
management of patients with cancer. Proc Nutr Soc. 2008; 67: 257-262
6. Forrest LM, McMillan DC, McArdle CS. Evaluation of cumulative prognostic scores
based on the systemic inflammatory response in patients with inoperable nonsmall cell
lung cancer. Br J Cancer 2003;89:1028-1030
7. Llobera J, Esteva M, Rifa J. Terminal cancer: duration and prediction of survival time.
Eur J Cancer 2000;36:2036-43)(for KPS)
8. Loprinzi CL, Laurie JA, Wieand HS, Prospective evaluation of prognostic variables from
patient- completed questionnaires. North Central Cancer Treatment Group. J Clin. Oncol
1994;12(3):601-7 (for ECOG)
9. Lau F Downing M, Lesperance M, Karlson N, Kuziemsky C, Yang J Using the Palliative
Performance Scale to provide meaningful survival estimates. J of pain and symptom
management 2009;38:134-144
10. Weng L, Huang H, Wilkie DJ, Hoenig NA, Suarez ML, Marschke M, Durham J.
Predicting Survival with the Palliative Performance Scale in a Minority-Serving Hospice
and Palliative Care Program
27
DRAFT DECEMBER 2009
Annex C - Survival Time by Age, Gender, Location, Diagnosis, and Initial PPS (Lau, 9)
No. of Patients Log
Variables Survival Times (in Days)
(%) Rank
Mean (95% Median (95%
Range P-value
CI) CI)
<1–
Overall 40 (37, 43) 8 (8, 9) 6066
2497
Age (years) 6063 <0.001
<1–
<45 74 (56, 92) 25 (18, 33) 280 (4.6%)
1199
<1–
45–64 45 (39, 51) 13 (11, 15) 1326 (21.9%)
1378
<1–
65–74 40 (34, 46) 8 (7, 9) 1525 (25.1%)
1633
<1–
75–84 38 (32, 43) 8 (7, 9) 1981 (32.6%)
1976
<1–
85+ 29 (21, 38) 5 (4, 6) 951 (15.7%)
2497
Gender 6038 <0.001
<1–
Female 47 (42, 52) 9 (8, 10) 3156 (52.2%)
2497
<1–
Male 33 (30, 37) 7 (6, 8) 2892 (47.8%)
1633
Location 6066 <0.001
<1–
Tertiary care 42 (35, 49) 10 (9, 11) 1264 (20.8%)
1976
<1–
Extended palliative care 45 (34, 56) 13 (11, 15) 710 (11.7%)
2497
<1–
Home crisis by PRT 39 (35, 43) 7 (6, 8) 4092 (67.5%)
1959
28
DRAFT DECEMBER 2009
No. of Patients Log
Variables Survival Times (in Days)
(%) Rank
Mean (95% Median (95%
Range P-value
CI) CI)
Diagnosis type 6058 <0.001
<1–
Cancer 39 (36, 42) 9 (8, 10) 5117 (84.5%)
2497
<1–
Noncancer 48 (37, 59) 4 (3, 5) 941 (15.5%)
1622
Diagnosis category 6066 <0.001
Cancer
<1–
Breast 64 (52, 75) 16 (12, 20) 556 (9.2%)
1092
Eye, brain, and central
37 (26, 47) 13 (9, 17) 1–480 134 (2.2%)
nervous system
Gastrointestinal
<1–
Colorectal 39 (31, 47) 11 (9, 13) 599 (9.9%)
1199
Esophagus 26 (4, 35) 7 (5, 9) <1–309 137 (2.3%)
<1–
Pancreas 34 (19, 49) 8 (6, 10) 294 (4.8%)
1959
<1–
Stomach 50 (26, 74) 11 (8, 14) 147 (2.4%)
1063
Other 36 (30, 41) 5 (3, 7) <1–400 154 (2.5%)
Genital—female 49 (37, 60) 17 (12, 22) <1–962 296 (4.9%)
<1–
Genital—male 53 (39, 67) 12 (8, 16) 369 (6.1%)
1305
Head and neck 22 (14, 30) 8 (6, 10) <1–177 79 (1.3%)
29
DRAFT DECEMBER 2009
No. of Patients Log
Variables Survival Times (in Days)
(%) Rank
Mean (95% Median (95%
Range P-value
CI) CI)
<1–
Hematopoietic 57 (28, 86) 6 (5, 7) 351 (5.8%)
2497
Respiratory 26 (23, 29) 8 (7, 9) <1–518 1255 (20.7%)
Skin 30 (19, 42) 9 (6, 12) <1–422 107 (1.8%)
<1–
Urinary 39 (27, 52) 10 (6, 14) 228 (3.8%)
1103
Other 27 (20, 34) 8 (6, 10) <1–968 391 (6.4%)
Noncancer
AIDS 62 (32, 93) 13 (5, 21) <1–652 75 (1.2%)
<1–
Cardiovascular 51 (32, 69) 3 (2, 4) 328 (5.4%)
1119
<1–
Neurological 53 (0, 109) 5 (3, 7) 69 (1.1%)
1622
<1–
Respiratory 40 (19, 60) 4 (3, 5) 157 (2.6%)
1111
Others 36 (14, 58) 4 (3, 5) <1–787 127 (2.1%)
<1–
Not recorded 54 (28, 81) 4 (3, 5) 213 (3.%)
1633
Initial PPS 6066 <0.001
PPS 10% 3 (1, 5) 1 (1, 1) <1–429 570 (9.4%)
PPS 20% 7 (4, 11) 2 (2, 2) <1–851 737 (12.1%)
<1–
PPS 30% 20 (16, 24) 5 (5, 5) 1420 (23.4%)
1274
<1–
PPS 40% 39 (34, 44) 13 (12, 14) 1647 (27.2%)
1119
30
DRAFT DECEMBER 2009
No. of Patients Log
Variables Survival Times (in Days)
(%) Rank
Mean (95% Median (95%
Range P-value
CI) CI)
<1–
PPS 50% 76 (64, 88) 28 (25, 31) 1055 (17.4%)
2497
PPS 60% 92 (80, 105) 43 (38, 48) 1–1199 487 (8.0%)
141 (92,
PPS 70% 63 (48, 78) 1–1959 150 (2.5%)
190)
Hazard Ratios for Age, Gender, Location, Diagnosis, and Initial PPSa (Lau, 9)
Variables Hazard Ratio 95% CI for Hazard Ratio P-value
Lower Upper
Age group (years) (vs. 85+) <0.001
<45 0.733 0.631 0.851 <0.001
45–64 0.830 0.757 0.909 <0.001
65–74 0.905 0.830 0.987 0.025
75–84 0.861 0.793 0.934 0.001
Gender (vs. male) 0.802 0.757 0.851 <0.001
Location (vs. home crisis by PRT) <0.001
Tertiary care 0.775 0.725 0.829 <0.001
Extended palliative care 0.725 0.666 0.790 <0.001
Diagnosis category (vs. cancer—respiratory) <0.001
Cancer
Breast 0.753 0.677 0.839 <0.001
31
DRAFT DECEMBER 2009
Variables Hazard Ratio 95% CI for Hazard Ratio P-value
Lower Upper
Eye, brain, and central nervous system 0.618 0.515 0.742 <0.001
Gastrointestinal
Colorectal 0.903 0.818 0.997 0.043
Esophagus 1.061 0.887 1.269 0.518
Pancreas 0.923 0.811 1.050 0.225
Stomach 0.802 0.672 0.956 0.014
Other 0.987 0.834 1.168 0.880
Genital—female 0.874 0.765 0999 0.048
Genital—male 0.654 0.578 0.740 <0.001
Head and neck 1.009 0.801 1.271 0.941
Hematopoietic 0.996 0.881 1.125 0.947
Skin 0.872 0.714 1.066 0.183
Urinary 0.754 0.653 0.869 <0.001
Other 0.885 0.789 0.994 0.039
Noncancer
AIDS 0.799 0.620 1.028 0.081
Cardiovascular 0.650 0.570 0.742 <0.001
Neurological 0.634 0.493 0.815 <0.001
Respiratory 0.884 0.746 1.048 0.157
Others 0.844 0.697 1.023 0.084
Not recorded 0.800 0.688 0.932 0.004
Initial PPS (vs. PPS 10%) <0.001
32
DRAFT DECEMBER 2009
Variables Hazard Ratio 95% CI for Hazard Ratio P-value
Lower Upper
PPS 20% 0.542 0.485 0.606 <0.001
PPS 30% 0.247 0.222 0.274 <0.001
PPS 40% 0.144 0.130 0.160 <0.001
PPS 50% 0.091 0.081 0.102 <0.001
PPS 60% 0.070 0.061 0.080 <0.001
PPS 70% 0.056 0.046 0.069 <0.001
a
This model is based on 6045 patient cases; 21 other cases were omitted because of missing data.
33
DRAFT DECEMBER 2009
Annex D
The Edmonton Symptom Assessment Scale (ESAS)
ESAS.pdf
M.D. Anderson Symptom Inventory (MDASI)
MDASI.pdf
Condensed MSAS (CMSAS)
CMSAS.pdf
The Cambridge Palliative Assessment Schedule ( CAMPAS-R)
CAMPAS-R.pdf
34
DRAFT DECEMBER 2009
Annex E
SLWG Group 3 Membership – Assessment Tools and Advance Care Planning
Sub Group*
John Welsh (Chair) 2
David Carroll 2
David Linden 2
Jackie Chaplin 2
Paul Baughan 1
Rosalie Dunn 1
Scott Davidson 1
Shirley Fife 1
Parvez Thekkumpurath 3
Fiona Wylie 3
Sandra White 3
Val Miller 3
Adam Gillespie 4
Bruce Mason 4
Christina Hamill 4
Sandra Campbell 4
Irene Stevens 4
Chris Sugden 5
Maria McGill 5
Tom Kane 5
Patricia Wallace
Shirley Byron
* 1 – Early identification, 2 – End of life, 3 – Tools for symptoms, 4 – Advance Care Planning, 5 – DNA
CPR (now being developed through eHealth group)
35
Related docs
Other docs by HC1207300654
