/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97163-en.cckm

201606168

page

100

UWHC,UWMF,

Tools,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

Palliative Performance Scale PPS - A New Tool

Palliative Performance Scale PPS - A New Tool - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


F
Palliative Perfoffnance Scale (PI5)’
A New Tool
FERN ANDERSON, G. MICHAEL DOWNING and JAN HILL, Victoria Hospice Society, and LYNN CASORSO
and NOREEN LERCH, Capital Region District Home Nursing Care, Victoria, British Columbia, Canada
5
F
3
OJ
z
(t)
’J
o
(J
k
k
Abstract / The Palliative Performance Scale (PPS), a
modification of the lGmobky Performance Scale, is pre
sented as a nerrr tool for measurement of phpiel status
in palliative care. lts infial uses in Vbtoda include commtr
nication, analysb of home nueing care worldoad, profiling
admissions and discharges to he hospice unit, and, pos
sibly, prognosticafnn. We assessed 119 palienb at home,
of whom 87 (tT/"|had a PPS raling betuveen 4ff/o utd
70%. At 213 patients admitted to the hospice unit, 175
(8/d were PPS2Oo/o-SV/o on admission. The average
period until death for 129 patients who died on the unit
was 1.88 days at 10% PPS upon admission,2.62 days
at 20o/o, 6.70 days at 30%, 10.30 days at 4Oo/", 13.87
days at 50o/o. Only two patients at 60% or higher died
in the unit. The PPS may become a basis for comparing
drug costs at home and lor studying the effects of treat-
ments (e.9. hypodermodysis) at various levels of physical
performance. Validity and reliability testing are cunently
being undertaken.
R6sum6/ L’6chelle d’indce fonclionnel, une variante de
l’6chelle Kamofsky, est pr6sent6e comme un nouvel
instrument pour 6valuer l’6tat physique des patients en
soins palliatifs. Initialement, l’6chelle a 6t6 utilis6e pour
6valuer le niveau de communication, la charge de tra-
vail associ6e aux soins du malade A domicile, le profil
des admissions et des cong6s de l’Unit6 et, enfin, son
utilit6 comme outil de pronostic. Parmi ’119 patients
soign6s A domicile, 87 (73 7o) d’entre eux avaient un
indice fonciionnel qui variait de 4OTo d 70 % selon
cette 6chelle. Des 213 patients admis A I’unit6 de soins
palliatifs, 175 (83 %) d’entre eux d6notaient, lors de leur
admission, un indice fonctionnel qui variait enlre 20 o/o
et 50 %. Enfin, 129 de ces patients sont deced6s er
I’h6pital et lzur du e moyenne a et6 respectivement de:
1,88 jours avec un indice fonctionnel de 10 o/oi & 2,62
jous et un indice de 20 "/"; de 6,70 jous et un indice de
30 % ; de 10,30 jous et un indice & 4O o/o i de 13,87 jous
et un indice d 50 o/o. Seuls deux patients, dont I’indice
6tait de 60 % et plus sont mortrs d I’Unit6. L1ndce
fonctionnel peut s’av6rer utile pour faire une
comparaison entre le coOt des m6dicaments pour les
soins A domicile et ceux i l’h6pital ; enfin, pour 6valuer
I’etficacit6 des traitements (i.e. I’hypodermoclyse) A
divers niveaux de I’indice fonctionnel. Des 6tudes ont
d’ailleurs 6t6 entreprises pour 6valuer la validit6 et la
fiabilit6 de cette echelle
INTRODUCTION
ln order to compare patients accurately within
and among palliative care programs, it is neces-
sary to have an assessment tool which reflects
the changing physical condition of palliative
care patients. Simple descriptors such as age,
disease, and gender allow for some demo-
graphic eyaluation. However, this becomes lim-
iting when one patient, matched for these factors
and fully ambulatory, is compared to another
who is bed-bound and moribund.
The Kamofrlg Perfomrance ftale (I(PS) (1), fir:st
developed in 1948, is a simple tool which has be n
used by several cancer agencies for many years. It
has since found use in other areas induding fr:nc-
tional status in cancer (2), rcnal disease (3), prognos-
tication (45,6), and gerontolory A.Odrer tools de-
signed for functional assessment indude Activitie of
Daily Living (ADL) (8), Instrumental Activitie of
Daily Living (IADL) (9) and Eastem Cooperative
Oncology Group @COC) scale (10).
More recently, acuity measurernent tools have
also been developed to reflect the severity of
symptoms. Examples of these tools are the Ed-
monton Symptom Assessment Scale (ESAS) (11)
and the San Diego Severity Index (SDSI) (12). A
recent artide by Degner and Sloan (13) found a
relationship between symptom distress at initial
diagnosis to prognosis in lung cancer. Other
workers have shifted to the development of qual-
ity of life tools which focus on at least four aspects
of life - physicaf functional psychological, and
social. As reviewed by de Haan et al. (14) in
studying stroke patients, there are many tools in
this category such as the Support Team Assess-
merrt Schedule (STAS) (15,1.6), COOP Charbs (17),
Mcfulaster Health Index Questionnaire (18), Not-
ti.gham Health Profile (L9), Spitzer Quality of
Life Index (20), Euroquol (21) and the Siclcress
Impact Profile (22). Cohen and Mount (23) have
discussed the challenge of developing an ad-
equate tool to measure quality in the palliative
patimt and have designed the McGill Quality of
Life Questionnaire (MQOL) (2a).
We present the Palliative Performance Scale
(PPS), a new tool based on the KPS, which guides
assessment of functional performance and pro-
vides a framework for measuring the progressive
decline in pailiative patients. In our view, the
,i
O\dfl
:)7
o{ ,^.HU
vo
.:q
dO
: ,ql
FX

t
I
Table 1 / PALLIATIVE PERFORMANCE SCALE ITTS)*
Anbulation Activity and
Evidence of Disease
Self{are Conscious
Level
Full
6
x
F
3
o
z
p
0l
6
E
0,,
I
(!
H
k
o
o
Normal Activity
No Evidence of Disease
Normal Activity
Some Evidence of Disease
Normd Activity wittr Effort
Some Evidence of Disease
Unable Normal loblWork
Some Evidence of Disease
Unable Hobby/House Work
Significant Disease
Unable to Do Any Work
Extensive Disease
As Above
As Above
As Above
As Above
Full
Full
Full
Occasional Assistance
Necessary
Considerable Assistance
Required
lvlainly
Assistance
Total Care
Total Care
Total Care
Normal
Normal
or Reduced
Normal
or Reduced
Normal
or Reduced
Normal
or Reduced
Normal
or Reduced
Reduced
Minimal
Sips
Mouth Care
Only
Fuil
Full
Full or
Confusion
Full or
Confusion
Full or Drowsy
or Confusion
Full or Drowsy
or Confusion
Full or Drowsy
or Confusion
Drowsy or
Coma
70
50
50
Reduced
Reduced
Mainly Sit/Lie
N,Iainly in Bed
Toally Bed
Bound
As Above
As Above
Death
* This scale is a modification of the Karnofskv Performance Scale. It takes into account ambulation, activiW, self-care, intake,
and conscious level.
10
Kamofsky scale suffers from being somewhat
outdated. For exa:nple, a rating of.3tr/o states hos-
pitalization is indicated artd 20Yo states it to be
necessary. Although that may have been the case
at one time, it does not adequately reflect the
"closer to home" (25) shift of resources in many
health care systems today. The PIrS contains no
reference to location of care. Rather, parameters
for assessment include factor related to physical
decline sudr as intake, mobility, and level of con-
sciousness. By more reflecerg physical
change, it is anticipated it can be used for progno6.
tication, researctr, and for program planning re-
lated to symptom control, drug costs, nursing and
auxiliary and respite care needs.
METHODS
Design of the Palliative Perforrrance Scale (PPS)
The Palliative Performance Scale (PPS) is shown
in Table 1 as it has been modified from the
Karnofsky scale, which can be seen for compari-
son in Table 2. Physical performance is divided
into 11 categories, measurd in 10% decremental
levels, from fully ambulatory and healthy (100%)
to death (0%). The factors which differentiate
these levels are based on five observable param-
eters: the degree of ambulation, ability to do ac-
tivities/extent of disease, abitty to do self-care,
food/fluid intake, and state of consciousness.
The tr(PS included three of these (ambulation,
activity/extent of disease, and self-care) but also
referred to the need for hospitalization, whidr fo-
cuses on location of care but does not he$ in
defining perfonnance and thus did not seem ap-
propriate to be included. Based on clinical experi-
ence with the dying, it was clear that an additional
two factors (intake and level of consciousness)
were necessary to adequately mecrilrre physical
performance. [r designing the PPS, the three K[’S
facto$ were maintained along with their respec-
tive % levels. Intake and consciousness levels
were then su$ectively scaled and added in a way
that reflects coruncn dinical situations. Based on
initial feedback by staff, some of the scaling has
been modified up or down to improve the "fit’.
As with KI usage and interpretation, the PIIS
Ievel for a given patient is determined by reading
across the table at eadr L0% decrement to find the
overall best fit Thus, using Table L, a palliative care
patient at the 30Plo level is totally bed-bor:nd" cmrnot
do any work, has extensive disease, requires total
physicalcare (i.e. cannotshave, domouth care, bathe,
etc.), has minimal intake, and is either fully conscious
or may be drowsy or confusd.
"Strongey’’ perfonnance factors are located on
the left and "softer" ones on the right. Judgment
is required if one the five factors observed in a
particular patient does not fit with the others. For
example, drowsiness (right side) can be seert in

7
Table 2 / KARNoFSKY PERFoRMANCE scALE (1)
Definition Criteria
Able to carry on normal activity and to work.
No special care is needed.
Unable to work Able b live at home, care for moet
p Nonat needs. A varying amount of assistance is reeded.
Unable to care for self. Requires equivalent of
institutional or hospital care. Disease may be
progressing rapidly.
Normal; no complaints; no evidence of disease
Able to carry on normal activity; minor signs
or symptoms of disease
Normal activity with efforb some signs or
symptoms of disease
Cares for seU. Unable to carry on normal
activity or to do active work
Requires occasional assistance, but is able to care
for most of his needs
Requires considerable assistance and frequent
medical care
Disabled; requires special care and assistance
Severely disabled; hospitalization is indicated
although death not imminent
Very sick; hospitalization necessary, active
supportive treatment necessary
Moribund; fatal processes progressing rapidly
Dead
7
F
3
z
50
(’l
q
h
k
OJ
100
90
80
60
rtO
30
10
0
patients at several levels but has less weight than
whether the patients can still walk or not The
usual approach to decide PffS is to first deter:nine
whether the disease has limited the ability to work
or ambulate (left side). If ambulation is limited, is
the person still able to care for himself/herselfl If
unable to do so, is s/he taking good or poor in-
take, and finally, has the level of consciowness
deteriorated and to what degree? Thus, the PPS
scale declines from top to bottom but also has left
to right implications.
MEASURING THE PPS
The Capital Regional District (CRD) in British
Columbia has eight home nursing care offices
that serve Victoria and the surrounding region.
Victoria Hospice and CRD Home Nursing Care
work in parbrership to provide pallinfivs care at
home. On average there are L50 registered pal-
liative care patients at any time, most of whom
remain at home. There are 6fi) deaths per year
including 300 home deaths. Services provided
include an interdisciplinary team of nurses,
counselors, physicians, volunteers, physiothera-
pists, and home support workers.
Home care rurrses record the PPS rating on
the in-home charts of all palliative care patients.
This rating is adjusted after each home visit.
Visits vary from biweekly to daily according to
need. The nurse then also records, on a pallia-
tive log kept at the home nursing care office, the
PPS value along with actual amount of direct
and indirect nursing care time spent. This in-
formation is entered into a computer database
for statistical analysis, induding numbers of pa-
tients at different PI’S levels, amounts of nursing
time spent with patients at various PPS levels,
comparisons between nursing offices of PIrS sta-
tistics, etc. For this paper, one method was to
analyze the profile of all registered patients on a
given day as recorded on the palliative lop of
each home nursing office. Other calculations were
based on the monthly accumulated amounts of
nursing time for patienb at each PPS level.
Victoria Hospice patients who require admis-
sion to acute care for assessment and manage-
ment are admitted to the hospice unit at the Royal
Jubilee Hospital. The PPS was initially recorded
Figure 1 / pw PRoFILE oF pALLLATtvE CARE
PATIENTS AT HOME
30
,3 20
t0
o. 15
I
Tro
G
s

8
F
3
z
(n
.’l
q
e
r
H
Figure 2 / }JOIvtr NURSING HOURS BY PPS RATING
200
a
h
5.
a 150
100
50
0
r Sep ’93
M Oct’93
e Nov’93
r Dec’93
50
PPS Level (%)
on admission and discharge only but is now done
on a daily tracking basis. Data here represent the
PIrS distribution at admission and discharge over
a one.year period. Prognosis resulb in this PaPer
were calculated for patients admitted to the hos-
pice unit during the same period using the admis-
sion date and PI’S compared to death date. Data
on prognosis for home patients are being collected
but are unavailable at this time.
RESULTS
Patients at Home
To look at the overall palliative care program case
profile, a "snapslrof’ dist$ution of palliative cale
patimb was taken on a specific day us illushabd in
Figure 1. The majority of the 1L9 palliative carc pa-
tients athome were in dre PIiS 4ry/r7tr/o range. I-ess
tharr one fiflh of patients were Pf ff/o ot below.
Figure 2 shows the amount of home nursing
hours spent in relation to the PPS level. Consist-
ent with the general PPS patient profile (Fig. 1),
the highest amount of nursing time is focused in
the midrange. However, there are somewhat
fewer hours spent with patients at the 70% level
and more at the 30% level.
Hospice Unit
Figure 3 slrows the PI profile for the hospice unit
Ior A3 admissions. Most patienb on adrdssion are
betweerrPl W’andtrh. Over onehalf of patients
died (PI 07.) in the r:rril The ronaining patienb
were discharged at PIjS levels behpeen 30elo xdtr/o.
The PI is being considered for use as a workload
measurement tool Unit mraes have found that, in
tle PPS 4ff/" to 30olo range, physical care denrands
ate hida As deatr ap’proadre in the PI 2Vh ao,td
Figure 3 / HOSPICE UMT PPS PROFILE ON
ADMSSION AND DISCHARGE
70 60 50 4A 30 20
PPS Level (%)
10pl" levels, heavy and timeduced, e.g. transfers to commode, Arjo tub baths,
time b assist intake, etc.
In Figure 4, data from the home profile (Fig-
ure L) and unit admission profile (Figure 3) are
combined to compare the populations. There is
some dilference between the two profiles, with
the unit having proportionately lower PPS
("sicker") patients. This is to be expected since
the hospice unit is intended primarily for assess-
ment and treatment of patients from home
whose symptoms are out of control and thus
presumably have lower PPS ratings.
Prognosis
Table 3 shows the length of life at varior:s PIS levels
for 213 patients who were admitted and L29 who
’J 50
6
+44
H
t30
o
ts<

v
HOME PATIENTS
oL5
ql
p20
Hts
FI
G
810
st
Figure 4 / Pfr LEVEIS FOR HOSPICE UNIT AND
ffi5040
PPS Level f/d
100
died in the hospice mit during 1993. All patients
admitted at]fP" PPS and 96% of those at20ol" P[5
died in the r:nit. The percentage of patients who
die in the r:nit decreases steadily from 100/oatPPS
L0"/o to 0o/o at Pffi 70o/o or higher. The average
numberof days from admission to death is strort-
est for those with the lowest admitting PPS, being
1.88 days at PPS 7V/", and increasing steadily to
13.87 days at PPS 50%. There is also a widening
range of values, with higher PIIS levels up from 13
days at PPS 10"/o b 34 days as Pf Sfflo. Also, at PI
ff/q Etro patients out of 20 died in tte r.urit in a
relatively slrort period of four da1a, indicatingmajor
corplications ad npid decline.
One unanticipated observation is that, of
those who were discharged alive from the unit,
the lurgth of stay was inversely related to the
PPS level beng 25 days at PPS 2oo/o and 8.29
days at PPS 70%.
DISCTJSSION
The Palliative Perfonnance Scale is in extensive
use in Victoria. While all care must be individu-
alized and care taken to not lose a person’s indi-
viduality by "categottzing" them, the use of the
PPS has proven an effective tool and a succinct
method of communicating in both verbal and
written forrrats. Rather than taking the time to
document the five categories within the PPS, the
simple statement "Mr. Jones is at PPS 20o/o" prv
vides much information including degree of
ambulation, amount of self-care, nutritional in-
take, and level of consciousness. In itse4 this
makes the PI’S appreciated and valued by home
care and hospice staff. In our experience, these
11 levels do clinically fit with commonly ob-
served patient physical and performance states.
It is a simple assessment which can be com-
pleted within minutes and can be learned
quickly.
Home nursing care uses the PPS in planning
visits and alerting staff to the need for increased
visit frequency or additional home support
worker hours. For example, a weekend home
visit is now usually planned for all patients who
ate 40o/o or lower. This reduces the number of
9
3
o
z
(D
o
d
I
(n
(J
(!
k
.ii
h
a6
(d
Table 3 / DlsposmoN AND LENGTH oF srAy, vIcToRIA HospIcE uNrr, 1993
A. Patients who Died in Unit
No. of
Pts.
PPS
on Admisaion Range% Die
Average
Days to
Death
10Y"
200h
30Y"
40o/o
50Y"
ffiYo
lWr
96%
82V"
%o/o
37%
l0o/o
1.88
2.62
6.70
10.30
73.87
4.00
6.56
0
1
o
15
11
4
I
2
5
8
11
4
113
0-11
1--26
7-28
4-38
F5
76
26
37
33
15
2
729
B. Patients Diecharged Alive from Unit
PPS
on Admiseion "/" Live
Average
Daye to
Death Range
No. of
Pte.
1
8
26
25
77
z
84
26
7
6
8
11
4
26
12.5
9.5
7
9
5
20Yo
30Yo
40o/"
50Yo
600/"
7jYo
3Yo
18o/"
44Y"
630/o
90Y"
700Yo
26.N
74.25
72.15
8.24
10.24
8.29
10.53
26
b27
1-60
1-20
2-34
}29

10
F
3
z
rn
o
(/)
(J
e
H
k
urgent and crisis situations which have to be
handled by weekend staff. Further, it is used to
teach family caregivers how to observe for and
communicate changes.in patient condition.
This study shows that palliative dients at the
PPS 70ol-100% level require little nursing time
beyond initial registration and introduction to the
palliative care program. As their disease
progresses, needs increase. During the PPS 30olo-
600/o rcrnge, services provided include coordina-
tion of services and supports, teaching client
caregivers to provide care, symptom m.rnage-
ment, and physician contacts. This "transition
time" is most difficult for patients; it is also where
CRD home c ue nurses spmd most of their time.
By the time the PPS is down to30"h or lower,
most services and supports have already been
organized. Careg"ivers are given support and
assistance in providing direct care. However, the
length of active intervention required from
home care nursing is reduced. This was an unex-
pected finding, as one might anticipate nursing
care to increase steadily as PPS declined. It may
be that local service factors, such as use of home
support workers (homemakers), volunteers, or
other services ofhospice account for the decreased
workload on nursing. It does, however, indicate
the need for further study and comparison with
other palliative care prograrns, especially from an
economic and efficiency viewpoint.
The PIrS has potential for use in prognostica-
tion. The initial findings here are based on rela-
tively small numbers. More extensive and de-
tailed collection is required as well as comparison
among palliative care populations before arry firrr
statements can be made. For example, the data
presented here in Table 3 are based on patients
admitted to our acute care hospice unit. It may be
that average lengths of life will be different possi-
bly longer for patients admitted to our new long-
tenn care palliative unit. Although detailed analy-
sis is not yet compiled on this unit or the home
palliative care patients, we have found the same
general picture for prognosis below PPS 30%. If
the PI can provide some accuracy even at the
lower levels, it will be of help in bed planning
workload planning, family information, death
plarning etc. A mudr larger database is currently
being developed to see iJ other factors such as
small data size, age, disease, or specific complica-
tions account for variations in these ranges indud-
ing the relation of the rate of decline to prognosis.
If so, the accuracy of the PIrS for prognostication
will be improved further.
Our data differ from those of Lassaunidre (26),
who reports the combined average time to death
at KPS 7tr/’-2ffi/" to be 16.8 days as compared to
our PPS levels of 1.88 and 2.52 days (2.3 days
combined). Of course these are differmt scales.
Nevertheless, more data analysis is critical. ln a
study by Boyd (27), reviewing short terrrinal ad-
missions at St. Christopher’s Hospice, 9.8Y. of
patienb died within 48 hours of admission. While
the overall purpose was to arraLyze the main prob’
lems and diagnosis of these patienb, it is difficult
to make a full comparison between our unit
(16.5"/, died within tt8 hours) and this study, since
in the latter no indicator of perfomrance level was
recorded.
The PIIS scale may prove valuable in research
into other aspects of patient care. When studying
clinical issues such as dyspnea, hypodermoclysis
(F{DC), drug costs in the home, confusion, new
analgesics, etc., perfomrance scales (28) are occa-
sionally use4 but overall little differentiation is
made in the patient population reported other
ttran age, gender, and disease (29,n\ For exam-
ple, it may be that use of HDC benefits patimts at
40"/o or higher PPS who are vomiting and con-
fused, but not those at lower levels. Are there dif-
ferences in costs of drugs or in the cost/benefit/
quality of freahent procedures at various PPS
levels?
By dividing the palliative care patient popula-
tion into several divisions, better answers may be
found than in grouping them all together regard-
less of extent of disease or imminence of death.
Using the ECOG performance status for ltrng can-
cer patients, Krech et al. (31) noted a general in-
crease in symptoms with worsening ECOG status
except for those with the lowest status who had
the fewest symptoms. It is possible that the PI’}S
scale which has lL rather than 4 levels may pro-
vide more detailed infor:rration.
Finally the relationship of several measurement
tools has not been studied. For example, a patient
at a particular PPS level may or may not be in
symptom distress (measured by scales such as
ESAS) and may or may not view their quality of
life as high (measured by quality of life scales). It
is suggested, then, that better trnderstanding of
advanced rlisease will occur when these three fac-
tors are studied in relation to eadr other. If so,
eventually palliative c:ue may be more accurately
reflected in a combined Performance/Distress/
Quality or "PDQ" scale.
In sum:nary, the PI’jS scale curently provides
valuable data for assessing, communicating, and
plarming care for palliative care patients in Victo-
ria, albeit in simple tenns. Validity testing of the
PPS is currently underway. It is presented at this
time to encourage its use in research, particu-
larly in viewing pailiative care patients not as
one homogmeous group but rather as a mixed

group in hansition from one state to another. It
is important that palliative care Programs recog-
nize such heterogenicity when comparing pa-
tients.
Date receiaed, February 14,1-994; date accepteil May
1"9, L994.
REFERENCES
L. Karnofsky DA, Albelman WH, Craver LF,
Burchmal IFL The use of nitrogen mustards in the
palliative treahnent of carcinoma. Cancer 1948; l:
5%4%.
2. Yates IW, Chalmer B, McKegney FP. Evaluation of
patients with advanced cancer using the Kamofsky
Performance Status. Cancer 1980;45: 22n-2224.
3. Evans RW, Manninen DL, Garrison LP et al. The
quality of life of patients with end stage disease. N
Engl J Med 1985;312:55F559.
4. Evans C, McCarthy M. Prognostic uncertainty in
terminal care: can Karnofsky lndex help? Lancet
1985 May; 1(8a39): 120F120e.
5. Pollack IF, Lunsford LD, Flickinger IC, Demeshek
HL. Prognostic factors in the diagnosis and treat-
ment of primary central nervous system
lymphoma. Cancer 7989; 53: 742:747.
5. Kemeny N, Niedzwiecki D, Shurgot B, Oderman P.
Prognostic variables in patients with hepatic
metastases from colorectal cancer. Importance of
medical assessment of liver involvement. Cancer
7989:63:792J97.
7. Crooks V Waller S, Smith T, Hahn TI. The use of
the Kamofsky Scale in determining outcomes and
risk in geriatric outpatients. J Gerontol l99l;46(41:
|/dt39-t4J.
8. Katz S, Ford AB, Moskowitz RW, fackson BA, Jaffe
ME, Cleveland MD. Studies of illness in the aged.
The lndex of ADL: a standardized measure of bio-
logical and psychological function. fAMA 1963;
185: 914-919.
9. Lawton MP, Moss M. Fulcomer J, Kleban MH. A
research and service oriented multilevel assess-
ment instrument. I Gerontol t982;37t 9749.
10. Zuhod CG Sctureiderrnan lvL Frei E et al. Appraisal of
methods for the study of of cancer in
maru comparative tial of nitnogen mustard
and triethylene thiophosphoramide. I Ckon Dis 1960;
tLtT-33.
11. Bruera E, Kuehn N, Miller MJ, Selmser P,
MacMillan K. The Edmonton Symptom Assess-
ment System (ESAS) for the assessment of pallia-
tive care patients. ] Palliat Care 1991;7(2\ 64.
12. Strause L, Herbst L, Ryndes T, Callaghan M, Piro
L. A severity index designed as an indicator of acu-
ity in palliative care. J Palliat Care t993;9(4): 11-15.
13. Degner LF, Sloan ]A. Symptom distress in newly
diagnosed ambulatory cancer patients as a predic-
tor of survival in lung cancer. J Pain Sympt Manag
1995 Aug; 70(6\ 423a31.
14. de Haan R, Aaronson N, Limburg M, Hewer RL,
van Crevel H. Measuring quality of life in stroke.
Stroke 1993 Feb; 24(21: 32W327.
15. Higginson I], McCarthy M. Effectiveness of two
palliative support teams. J Public Health Med 1992
Mar; 14(1): 5G-56.
15. Higginson I], McCarthy M. Validity of the support
team assessment schedule. Palliat Med 1993;7(3):
27W278.
1.7. Nelson E, Wasson L Kirk J, Keller A, Clark D,
Dietrich A, Stewart A, Zubkoff M. Assessment of
function in routine clinical practice: description of
the COOP chart method and preliminary findings.
I Chron Dis 1987; 40 (suppl 1): 555-635.
18. Chambers LW, Sackett DL, Goldsmith CH,
MacPherson AS, McAuley RG. Development and
application of an index of social function. Health
Serv Res 7976;11: 430-41.
19. Hunt MS, McKenna SP, McEwen L Backett EM,
Williams I, Papp E" A quantitative approach to per-
ceived health status: a validation study. J
Epidemiol Community Health 1980; 34: 281,-285.
20. Spitzer WO, Dobson AJ, Hall ]. Chesterman E, Levi
J, Shepherd R, Battista RN, Catchlove RR. Measur-
ing the quality of life of cancer patients: a concise
Ql-index for use by physicians. J Chron Dis 1981;
34t 585-597.
21. The Euroquol Group. Euroquol: a new facility for
the measurement of health-related quality of life.
Health Policy 1990; 19:199-208.
22. Bergner M, Bobbitt RA, Carter WB, Gilson BS. The
sickness impact profile; development and final re-
vision of a health status measure. Med Care 198L;
19:787-805.
23. Cohen S& Mount BM. Quality of life in terminal
illness: defining and measuring subjective well-be’
ing in the dyi"g. I Palliat Care 1992:8(3): 4045.
24. Cohen SR, Mount BM, Strobel MG, Bui F. The
McGill Quahty of Ufe Questionnaire: a measure of
quality of life appropriate for people with ad-
vanced disease. A preliminary study of validity
and acceptability. Palliat Med 1995: 9(3):207-219.
25. Seaton PD, et al. Closer to home - Report of the
British Columbia Royal Commission on Health
Care and Costs. Victoria: British Columbia Minis-
try of Health, 1991.
26. Lassaunilre IM, Vinant P. Prognostic factors, sur-
vival, and advanced cancer. J Palliat Care 7992;
8(4): 52-54.
27. Boyd K|. Short terminal admission to a hospice.
Palliat Med t993; 7: 289+94.
28. Fhn IW, Walsh TD, MacDonald N, Bruera E, Krebs
LU, Shepard KV. Placebo blinded study of mor-
phine sulphate sustained-release tablets and imme-
diate-release morphine sr’lFhate solution in outpa-
tients with chronic pain due to advanced cancer. J
Clin Oncol 1993 May; L1(5):967-972.
29. Mercadante S, Spoldi E; Caraceni A, Maddaloni S,
Simonetti MT. Octreotide in relieving
gastrointestinal symptoms due to bowel obstruc-
tion. Palliat Med 1993; 7:295499.
30. Famcombe M, Chater S. Case studies outlining use
of nebulieed molphine for patients with end-stage
chronic lung and cardiac disease. ] Pain Sympt
Manag 1993; 8(4): 221.-225.
31. Krech RL, David J. Walsh D, Curtis EB, Symptoms
of lung cancer. Palliat Med 7992; 6:309-315.
L1
F
3
0.)
z
(,
(n
I
e

L