Embed Size (px)
Citation preview
ORIGINAL ARTICLE
Quality of Life of the Malaysian General Population:Results from a Postal Survey Using the SF·36
£ a.
A B Azman, PhD*, S Sararaks, MPH*, B Rugayah, MPH**, L L Low, B.Se (Hons.)***, A A Azian, MPH****,S Geeta, MBBS****, C T Tiew, B.Se (Hons.)*****
'Public Health Specialist, Institute of Public Health, 50590 Kuala Lumpur, "Public Health Specialist, Clinical Research Centre,Hospital Kuala Lumpur, 50586 Kuala Lumpur, "'Research Officer, Institute of Public Health, 50590 Kuala Lumpur, ....Medical
Officer, Institute of Public Health, 50590 Kuala Lumpur, ....·Senior Statistician, Institute of Public Health, 50590 Kuala Lumpur
Introduction
Quality of life (QOL) emerged from sociology andsocial psychology as a global concept encompassingaspects of physical, social, emotional and spiritual wellbeing. Defined in terms of cultural standards andnorms, it has reference to the desires, needs,experiences and aspirations of the individual. Whenconsidered as a dimension or domain of quality of life,health is best thought of in the narrower sense offactors· that are generally considered to fall under thepurview of health care providers, or that are likely tobe the target of a health care intervention. Thus, theterm "Health-related quality of life (HRQOL)" hassometimes been preferred to that of QOL'·2 and this canbe defined as an individual's satisfaction or happinesswith domains of life as far as they affect or are affectedby health. It can be differentiated from QOL in thatHRQOL concerns itself primarily with those factors that
fall under the purview of health care providers andhealth care systems 3.
Instruments used in the measurements of bothindividual and population HRQOL are either generic,i.e. not specifically designed for patients with aparticular disease or condition, or, they may be specificfor a particular disease or condition but not applicableto the general population. We had chosen to use theSF-36 (Short form - 36), a generic outcome measure ofsickness 4. It is based on 36-items selected to representeight health concepts (physical, social and rolefunctioning, mental health, health perceptions, energy,fatigue, pain and general health)4. and shown to be asensitive measure for numerous diseases, thus its use inthe measurement of outcome of care 4.5. In addition, ithas been found to be sensitive to changes in health ingeneral populations 6". The developers have also
This article was accepted: 23 April 2003Corresponding Author: Azman Abu Bakar, Health Systems Research Division, Institute of Public Health, 50590 Kuala Lumpur
694 Med J Malaysia Vol 58 No 5 December 2003
Quality of life of the Malaysian General Population: Results from a Postal Survey Using the SF-36
methodically documented the validity and reliability ofthis instrument 8,9,10,11.
One of the many applications of HRQOL research is inmaking empirical statements concerning communitylife, urban development and social progress. Thisinformation will be of relevance to policy makers andhealth planners to improve the health services.However, in order to make the correct inferences inrelation to the population being studied, a referencepoint will be needed as a yardstick for comparison.
Normative data are the key to determining whether agroup or an individual scores below or above theaverage for their country, age or sex4
,l2, Although therealready exists published norms for the United States"the Queensland region and Australian Capital Territoryof Australia7,13, the United Kingdom" and Canadal2
,
comparable norms do not yet exist for Malaysia. In arecent study done locally, norms for the US generalpopulation were used as a guide for reflecting deficitsin the various domains because of disease, Le.comparing the general population functioning with thatof the diseased population15 , Given the cultural andethnic differences, not to mention the geneticdifferences, it is expected that the perceived HRQOL ofthe general population in Malaysia might differ fromthat of the general population of other countries.
Here, we present the population norms for Malaysia forSF-36 by age, sex and ethnicity,
Material and Methods
A nationwide household survey was conductedthroughout Malaysia with respondents who wererandomly selected from living quarters sampled for theThird Round of the Labour Force Survey# in 2000. Amultistage stratified (by state and urban/rural location)random sampling was done, proportionate to
population size@. The sampling design excluded thisinstitutionalised population that constituted less than3%16.
In total, 1,746 Enumeration Blocks' (EBs) werecanvassed, Within each EB, six questionnaires wererandomly given to three living quarters. A livingquarters" (LQ) in the EB. Respondents must beMalaysians, aged 18 years and above and literate ineither Bahasa Malaysia (BM) or English.
The enumerators for the Third Round of the LabourForce survey handed survey materials consisting of anintroductory letter with accompanying explanation onthe study, a bilingual questionnaire and a stampedreturn envelope to respondents by hand, in efforts toincrease response rates. Enumerators explained torespondents that their responses were important to theMinistry of Health, and that all answers would be keptconfidential. Respondents were advised to fill thequestionnaires and subsequently post them back to theresearchers as soon as possible. Response by proxywas not entertained.
The UK version of SF-36 had been translated to BahasaMalaysia by a group of researchers from University ofScience Malaysia##, A research team under the' aegisofInternational Quality of Life Assessment (IQOLA)Project had developed a translated version. We haveadapted and modified that translated version of SF-36.In-depth interviews were conducted for cognitivedebriefing on patients and their families attendinggovernment Medical Outpatients Clinics, with quotasampling to cover diverse ethnic groups. We exploredtheir perception, understanding and interpretation oftranslated items of SF-36 (BM version). Results of thecognitive debriefing were used to refine the translatedBM SF-36 questionnaire further and this was used in thesurvey. In the process of printing, typological errorswere noted in the questionnaire for questions 6 and 8which could affect the accuracy of responses for social
# The Labour Force Survey is a survey conducted by the Department of Statistics, mainly to measure the employment statusin the country. The sample selected for the survey is spread over quarterly rounds, in this case, the third quarter of 2000.
@ First stage of the sampling involved the selection of Enumeration Blocks (EB); while in the second stage of sampling livingquarters" (LQ) were selected.An enumeration block is an artificially created contiguous geographical area with specific boundaries (either natural orartificial) that do not straddle administrative boundaries. On average, an EB covers about 100-120 living quarters.A living quarters (LQ) is a living unit, structurally separate (surrounded by walls, fences etc. and is covered by a rn(1f\ :I ndindependent (has direct access via a public staircase, communal passages or landing) and are meant for living.
## A research team under the aegis of International Quality of Life Assessment CIQOLA) Project had developed a tl"lnsbt:,'dversion,
Med J Malaysia Vol 58 No 5 December 2003 695
ORIGINAL ARTIClE
functioning (SF) and bodily pain (BP) subscales.Question 6 had one response category left out("moderately") while question 8 had six responsecategories instead of the specified five. Independentraters gave levels/scores for each response category toidentify which categories should be grouped togetherto reduce the number of response categories from 6 to
5. "Mild" and "very mild" were combined together toform one category.
Calculation of scores for all the eight domains##followed that outlined by Ware et at'.
Data collection was carried out simultaneouslythroughout the country in September 2000. Theresearchers received questionnaires in the mail until 31December 2000.
Results
Out of 10,041 questionnaires sent out, 3072 returned ina usable form. Response rate was 30.6%.
The male to female ratio was 1.04. Mean age was 39.8± 12.9#; median was 39.4 and age in the sample rangedfrom 18 to 87 years. The majority were Malays (77.2%),followed by Chinese (16.4%). More than half (60.0%) ofthe sample was from urban areas.
Half (54.1%) of respondents had secondary leveleducation, with 27.3% with primary education. 61.8%were employed, and the reported average monthlyincome was RMI065 ± 1224", with a median of RM781and a range of RM20 to RM20,000. Two thirds (77.6%)of the sample owned some form of transport,commonly a motorcycle or a car; whilst half (55.5%)owned the house they were staying in at the time ofsurvey. Almost all (94.7%) reported to be staying withtheir family, with only 2.4% staying on their own, and2.9% with friends.
On self-reported morbidity, 26.0% reported some type
of disease, and 2.7% some form of handicap,commonest being complaint pertaining to therespiratory tract and musculoskeletal disorders.
Table I shows the characteristics of respondentscompared to the Malaysian population in 2000'6.
Urban dwellers, young males, Chinese and otherethnicity were under-represented.
Tables II to VI show the means, summarised percentileproportions together with floor@ and ceiling"" effectsfor all the 8 domains by age, sex, ethnicity and by ageby sex.
Females had lower means for all domains compared tomales, with the difference significant for PF, BP, VT,REE and MH. This picture is consistent, even afterstratifying by age.
As expected, increasing age was associated with areduction in mean PF for both gender, while mean REP,BP, VT and SF dropped from 60 years and above, andmean GH dropped a decade earlier, from 50 yearsonwards. REE and MH showed no obvious pattern withage, though young adults less than 30 years had lowermeans for both the domains.
Indians have lower means for PF, BP, VT, SF and MHcompared to other ethnic groups, though the differencewere not significant. Other Bumiputera had lowermeans only for BP, the level for which was similar tothat of Indians.
In general, ceiling effects were seen for PF, REP, SF andREE. Older population greater than 60 years hadminimal ceiling effects for PF, as expected. Flooreffects were not much of a problem except for REE.Figures 1 to 3 show the variability of the subscales byage, gender and ethnicity. Each of the subscalesdemonstrates a similar distribution for gender andethnicity. PF shows greater variability with increasingage. BP, GH, VT and MH do not show much difference
## Also referred to as subscales4• They are: physical functioning (PF), social functioning (SF), physical role functioning (REP),
emotional role functioning (REE), mental health (MH), energy fatigue/vitality (VT), bodily pain (BP) and general health(GH).
# This is one standard deviation. 95% confidence limits for the mean were 39.4, 40.3 years.This is one standard deviation. 95% confidence limits for the mean were RM1009.8, 1120.3.
@ Proportions of subjects receiving the maximum possible score.Proportions of subjects receiving the minimum possible score.
696 Med JMoloysio Vol 58 No 5 December 2003
Quality of Life of the Malaysian General Population: Results from a Postal Survey Using the SF-36
in distribution of scores across age, with all agesdemonstrating minimal floor and ceiling effects.
Figure 4 compares the Malaysian average# with that forthe general population of United States (US)" theAustralian Capital Territory (ACT) 7 and Canada12
. PFand VT for Malaysians were higher than the averagescores for the general population of US, though it wassimilar to that for Canadians. Malaysians had the lowestscores for BP and GH, significantly lower than the other
# Mean score with 95% confidence limits.
countries'. SF, REE and MH, though comparable to thenormative data for US, were lower than that for Canada(for the three subscales), and the ACT (for the last twosubscales). Only REP was found to be similar across allthe countries compared. Although the confidenceintervals do not overlap, the differences mentioned aresmall, with only VT, BP and GH with a gap of 5 or morepoints, the level considered to be clinically and sociallymeaningful 17.
Table I: Comparison of characteristics of respondents with Malaysian Population (Census 2000)
Characteristics Respondents Population Test of proportionsNumber % % p value
Area Urban 1832 60.0 62.0 0.04
(n=3055) Rural 1223 40.0 38.0 0.08
Sex Male 1563 51.1 51.0 0.48
(n=3061) Female 1498 48.9 49.0 0.48
Age group 18 - 29 years 262 17.0 29.0 0.03
- All Males 30 - 39 years 421 27.3 26.0 0.10
(n=1543) 40- 49 years 429 27.8 21.2 0.01
50 - 59 years 260 16.9 12.7 0.37
60 - 69 years 128 8.3 7.0 0.45
> 70 years 43 2.8 4.0 0.39
Age group 18 - 29 years 445 30.1 29.3 0.01
-All Females 30 - 39 years 455 30.8 26.3 0.09
(n=1479) 40 - 49 years 347 23.5 20.4 0.02
50 - 59 years 150 10.1 11.8 0.42
60 - 69 years 64 4.3 7.4 0.51
> 70 years 18 1.2 4.8 0.57
Ethnic Group Bumiputeraa 2373 77.2 65.1 0.000
(n=3072) Chinese 505 16.4 26.0 0.000
Indian 165 5.4 7.7 0.17
Others 29 0.9 1.2 0.001
Med J Malaysia Vol 58 No 5 December 2003 697
0-
0-0
Tabl
eII:
Nor
ms
for
the
Gen
eral
Mal
aysi
anPo
pula
tion,
Tota
lSam
ple
(Xl
;>0
G)
Tota
lSam
ple
(N=3
072)
PFRE
PBP
GH
VTSF
REE
MH
z » r-M
ean
85.9
882
.03
69.9
666
.74
66.7
983
.73
79.2
374
.66
» ;>0
25th
Perc
entil
e80
.00
75.0
062
.00
52.0
055
.00
75.0
066
.67
64.0
0--
-I n50
thPe
rcen
tile
(Med
ian)
95.0
010
0.00
72.0
067
.00
70.0
087
.50
100.
0076
.00
r-
rn
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
88.0
0St
dD
evia
tion
17.9
132
.12
17.5
919
.99
17.6
819
.28
35.9
217
.19
Ran
ge10
0.00
100.
0090
.00
100.
0010
0.00
100.
0010
0.00
100.
00%
Cei
ling
32.3
70.4
02.
44.
043
.271
.39.
2%
Floo
r0.
280
.20.
30.
30.
10.
212
.70.
0V
alid
NN
=30
72N
=30
64N
=30
70N
=30
70N
=307
1N
=30
70N
=30
60N
=307
1
Fem
ale
(N=
1498
)PF
REP
BPG
HVT
SFRE
EM
HM
ean
84.5
281
.47
68.9
666
.03
65.1
082
.94
76.9
273
.30
25th
Perc
entil
e75
.00
75.0
062
.00
52.0
050
.00
75.0
066
.67
60.0
050
thPe
rcen
tile
(Med
ian)
90.0
010
0.00
72.0
067
.00
65.0
087
.50
100.
0076
.00
75th
Perc
entil
e10
0.00
100.
0084
.00
82.0
080
.00
100.
0010
0.00
88.0
0St
dD
evia
tion
18.5
232
.55
17.5
620
.15
17.5
419
.55
37.2
517
.63
Ran
ge10
0.00
100.
0090
.00
100.
0095
.00
100.
0010
0.00
100.
00%
Cei
ling
27.8
69.5
02.
42.
440
.968
.48.
1%
Floo
r0.
18.
60.
30.
30
0.2
14.1
0.1
s:V
alid
NN
=14
98N
=14
95N
=14
97N
=14
96N
=14
98N
=14
97N
=14
92N
=14
98(J
)Q
...
'- s:
Mal
e(N
=15
63)
PFRE
PG
HVT
SFRE
EM
HQ
BPQ ~
Mea
n87
.38
82.4
870
.91
67.3
968
.46
84.4
881
.37
75.9
9o· ~
25th
Perc
entil
e80
.00
75.0
062
.00
52.0
055
.00
75.0
066
.67
64.0
00
,50
thPe
rcen
tile
(Med
ian)
95.0
010
0.00
74.0
067
.00
70.0
087
.50
100.
0076
.00
(Xl z
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
88.0
00
Std
Dev
iatio
n0
,17
.23
31.7
817
.56
19.8
217
.68
19.0
134
.51
16.6
40
Ran
ge10
0.00
100.
0090
.00
100.
0010
0.00
100.
0010
0.00
84.0
0(J
)() (J)
%C
eilin
g36
.771
.20
2.4
5.6
45.5
74.1
10.2
3 CJ
~%
Floo
r0.
37.
80.
30.
40.
10.
111
.30
i'.)
Val
idN
N=
1563
N=
1558
N=
1562
N=
1563
N=
1562
N=
1562
N=
1557
N=
1562
0 0 w
~Ta
ble
III:
Nor
ms
for
Age
Gro
ups,
Gen
eral
Mal
aysi
anPo
pula
tion,
Fem
ales
&M
ales
Com
bine
d.CD Q
...
'- ~
Ages
18-2
9PF
REP
BPG
HVT
SFRE
EM
HQ
Fem
ale
&M
ales
(N=7
09)
Q '< enM
ean
91.1
684
.96
1.93
70.1
866
.45
83.2
974
.29
72.3
7o' ~
25th
Perc
entil
e90
.00
75.0
062
.00
57.0
055
.00
75.0
033
.33
60.0
00
150
thPe
rcen
tile
(Med
ian)
95.0
010
0.00
74.0
072
.00
66.6
787
.50
100.
0072
.00
co z75
thPe
rcen
tile
100.
0010
0.00
90.0
082
.00
80.0
010
0.00
100.
0084
.00
0 01
Std
Dev
iatio
n13
.66
28.5
116
.61
17.4
016
.40
18.7
138
.25
17.1
60
Ran
ge10
0.00
100.
0090
.00
100.
0090
.00
100.
0010
0.00
100.
000
CD ()
c:CD
%C
eilin
g44
.472
.00
3.0
2.0
40.2
64.4
6.6
Q...
3..z
0-
%Fl
oor
0.1
5.2
0.1
0.1
00.
115
.40.
1~
Q...
IVV
alid
NN
=70
9N
=70
8N
=70
9N
=70
9N
=70
9N
=70
9N
=70
8N
=70
9r-
0~
0 wAg
es30
-39
PFRE
PBP
GH
VTSF
REE
MH
Q...
Fem
ale
&M
ale
(N=8
77)
3"-
<D
Mea
n89
.14
83.6
070
.94
67.8
167
.57
84.0
880
.25
75.1
7~
25th
Perc
entil
ec
85.0
075
.00
62.0
052
.00
55.0
075
.00
66.6
764
.00
0"
50th
Perc
entil
e(M
edia
n)95
.00
100.
0074
.00
70.0
070
.00
87.5
010
0.00
76.0
0~ c' ::::
:l
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
88.0
0G
)
Std
Dev
iatio
n14
.87
31.5
417
.07
19.0
917
.30
18.5
435
.74
16.4
8<D ::::
:l<D
Ran
ge95
.00
100.
0090
.00
95.0
095
.00
87.5
010
0.00
80.0
0a ""
"U
%C
eilin
g37
.474
.10
3.0
4.3
42.6
73.4
8.3
0 -a%
Floo
r0
7.8
0.1
00
012
.70
c: 0"
Val
idN
N=
877
N=
875
N=
876
N=
876
N=
876
N=
876
N=8
71N
=87
6g 2
Age
s40
-49
PFRE
PBP
GH
VTSF
REE
MH
;:oc
<D enFe
mal
e&
Mal
es(N
=781
)c: ur
Mea
n86
.27
83.9
769
.62
68.1
268
.38
85.4
081
.45
76.3
9Cf
25th
Perc
entil
e80
.00
75.0
062
.00
55.0
055
.00
75.0
066
.67
64.0
03 c
50th
Perc
entil
e(M
edia
n)90
.00
100.
0072
.00
70.0
070
.00
87.5
010
0.00
80.0
0""
"U 0
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
89.3
3en c
Std
Dev
iatio
n16
.51
29.5
417
.42
19.3
717
.86
17.8
834
.33
16.7
9C
/lc:
Ran
ge10
0.00
100.
0090
.00
100.
0090
.00
87.5
010
0.00
80.0
0:< ~
%C
eilin
g28
.871
.00
2.3
5.4
47.3
74.0
11.1
c en
%Fl
oor
0.3
6.0
0.1
0.1
00
10.9
0S
'co
Val
idN
N=7
81N
=77
8N
=78
0N
=781
N=7
81N
=78
0N
=77
8N
=781
3"-
<D C/l
0-
~'0 '0
0-
'IA
ges
50-5
9PF
REP
BPG
HVT
SFRE
EM
H0
0 0:-
0
Fem
ale
&M
ales
(N=4
11)
G)
Mea
n80
.10
79.2
169
.11
62.5
666
.31
83.7
081
.83
74
.85
z »25
thP
erce
ntile
70
.00
75
.00
62
.00
45.3
35
0.0
07
5.0
06
6.6
760
.00
r- »
50th
Per
cent
ile(M
edia
n)85
.00
100.
007
2.0
062
.00
70
.00
87
.50
100.
007
6.0
0:-
0--
l()
75th
Per
cent
ile95
.00
100.
009
0.0
080
.33
80
.00
10
0.0
010
0.00
92.0
0r-
rn
Std
Dev
iatio
n20
.73
34.5
11
7.8
722
.16
17
.80
20
.30
33.8
01
8.1
7R
ange
100.
0010
0.00
90
.00
100.
009
5.0
010
0.00
100.
0080
.00
%C
eilin
g2
1.7
67.6
01
.74
.44
5.3
73.6
9.7
%Fl
oor
0.2
10
.70
.21.
20
0.5
10.8
0V
alid
NN
=411
N=4
10
N=4
11N
=411
N=4
11N
=411
N=
40
9N
=411
Age
s60
-69
PFRE
PBP
GH
VTSF
REE
MH
Fem
ale
&M
ales
(N=
192)
Mea
n71
.51
67.5
863
.36
57.0
161
.98
80.2
178
.30
73
.83
25th
Per
cent
ile5
5.0
025
.00
50
.17
40.8
35
0.0
06
2.5
06
6.6
76
0.0
050
thP
erce
ntile
(Med
ian)
75
.00
100.
006
2.0
06
0.0
060
.00
87.5
010
0.00
76
.00
75th
Per
cent
ile90
.00
100.
0084
.00
72
.00
75
.00
100.
0010
0.00
88.0
0St
dD
evia
tion
22.2
94
1.2
52
0.3
921
.80
17.8
12
1.7
937
.02
17.7
3R
ange
90.0
010
0.00
90
.00
92.0
010
0.00
10
0.0
010
0.00
84.0
0%
Cei
ling
9.4
56.3
00
2.1
36.5
71.4
8.9
%F
loor
018
.81.
60
0.5
0.5
13.5
0V
alid
NN
=l9
2N
=l9
2N
=1
92
N=1
91N
=1
92
N=
l92
N=
19
2N
=1
92
Age
s70
&ov
erPF
REP
BPG
HVT
SFRE
EM
H~
Fem
ale
&M
ales
(N=6
1)CD Q
...
Mea
n6
4.3
467
.62
62
.84
52.0
556
.23
72
.75
73.2
271
.41
'- ~
25th
Per
cent
ile4
5.0
01
6.6
74
8.0
035
.00
40
.00
62
.50
33.3
35
4.6
70 0"
50th
Per
cent
ile(M
edia
n)68
.33
100.
006
2.0
04
8.0
05
1.6
77
5.0
010
0.00
69.3
3-< V
> o'75
thP
erce
ntile
90
.00
100.
007
9.0
07
2.0
078
.33
10
0.0
010
0.00
92.0
0~
Std
Dev
iatio
n26
.62
43
.38
19
.74
25.8
925
.06
9.2
741
.19
20.3
3tJ
'lR
ange
100.
0010
0.00
90
.00
97.0
010
0.00
100.
0010
0.00
84.0
0ex
> Z%
Cei
ling
4.9
60
.70
06.
636
.167
.216
.40 tJ
'l%
Flo
or3.
32
3.0
1.6
4.9
1.6
1.6
19.7
00 CD
Val
idN
N=6
1N
=61
N=6
1N
=61
N=6
1N
=61
N=6
1N
=61
n CD 3 0-
~ 1'0
0 0 w
~Ta
ble
IV:
Nor
ms
for
Fem
ales
byA
geG
roup
.CD 0
-<
-
Age
s18
-29
Fem
ale
(N=4
45)
PFRE
PBP
GHVT
SFRE
EM
H~ Q
Mea
n90
.44
85.6
271
.99
70.1
565
.61
82.9
874
.25
72.1
60 '< '"
25th
Perc
entil
e85
.00
75.0
062
.00
57.0
055
.00
75.0
033
.33
60.0
0c' ~
50th
Perc
entil
e(M
edia
n)95
.00
100.
0073
.33
72.0
065
.00
87.5
010
0.00
72.0
0tJ'
I75
thPe
rcen
tile
100.
0010
0.00
90.0
082
.00
80.0
010
0.00
100.
0084
.00
ex> Z
Std
Dev
iatio
n13
.11
28.2
915
.85
17.2
916
.30
18.5
038
.27
17.2
80 tJ
'IR
ange
80.0
010
0.00
90.0
085
.00
90.0
087
.50
100.
0010
0.00
0 CD%
Cei
ling
38.9
73.7
02.
51.
638
.964
.46.
10
nc
CDQ
...3
%Fl
oor
05.
20.
20
00.
415
.30.
2cr
~~
Val
idN
N=
445
N=
445
N=4
45N
=44
5N
=44
5N
=44
5N
=44
4N
=44
5g.
,N
c:
0A
ges
30-3
9Fe
mal
e(N
=455
)PF
REP
BPGH
VTSF
REE
MH
&"'
0 wg.
,M
ean
87.0
182
.65
69.6
566
.44
65.7
683
.90
78.1
074
.09
:T-
25th
Perc
entil
e80
.83
75.0
062
.00
52.0
050
.83
75.0
066
.67
63.3
3CD
50th
Perc
entil
e(M
edia
n)90
.00
100.
0072
.00
67.0
065
.00
87.5
010
0.00
76.0
0~ 0
-75
thPe
rcen
tile
100.
0010
0.00
90.0
082
.00
80.0
010
0.00
100.
0088
.00
'< en c'
Std
Dev
iatio
n16
.04
32.6
517
.24
19.2
717
.37
18.4
137
.27
16.8
2::
l
G)
Ran
ge95
.00
100.
0090
.00
95.0
095
.00
87.5
010
0.00
80.0
0CD ::
lCD
%C
eilin
g28
.873
.30
2.9
2.6
41.1
71.0
7.5
a%
Floo
r0
8.6
0.2
00
014
.40
-u 0 ""U
Val
idN
N=4
55N
=45
4N
=45
5N
=45
4N
=45
5N
=45
5N
=45
2N
=45
5c 0
-
Age
s40
-49
Fem
ale
(N=3
47)
PFRE
PBP
GHVT
SFRE
EM
Hg ::
l
Mea
n83
.37
81.0
766
.57
66.3
165
.76
83.8
577
.94
74.2
7;;
0CD en
25th
Perc
entil
e75
.00
75.0
052
.00
52.0
050
.00
75.0
066
.67
60.0
0c VI
"
50th
Perc
entil
e(M
edia
n)90
.00
100.
0062
.00
67.0
065
.00
87.5
010
0.00
74.6
73' 3
75th
Perc
entil
e95
.00
100.
0084
.00
82.0
080
.00
100.
0010
0.00
92.0
0c
Std
Dev
iatio
n17
.42
31.6
118
.45
20.0
318
.38
18.8
436
.22
18.1
8-u 0 en
Ran
ge95
.00
100.
0090
.00
100.
0090
.00
87.5
010
0.00
80.0
0a- en
%C
eilin
g22
.566
.50
2.3
3.5
44.8
68.8
10.7
c j%
Floo
r0
7.8
0.3
0.3
0.3
012
.70
cV
alid
NN
=34
7N
=34
6N
=34
6N
=34
7N
=34
7N
=34
6N
=34
6N
=34
7en ::;
'cc :T
-CD en
'-.l
--n
0W 0
-
"A
ges
50-5
9Fe
mal
e(N
:150
)PF
REP
BPGH
VTSF
REE
MH
00
;;0
tvQ
Mea
n73
.93
73.9
966
.67
58.9
362
.50
80.1
778
.52
72.7
5Z
25th
Perc
entil
e62
.08
50.0
056
.17
45.0
050
.00
67.7
166
.67
60.0
0» .....
..50
thPe
rcen
tile
(Med
ian)
80.0
010
0.00
68.6
758
.50
62.5
087
.50
100.
0076
.00
» ;;0
75th
Perc
entil
e91
.25
100.
0084
.00
77.0
075
.00
100.
0010
0.00
88.0
0--
-I n .......
Std
Dev
iatio
n23
.74
37.8
818
.73
22.8
317
.93
23.3
636
.57
19.0
3en
Ran
ge10
0.00
100.
0090
.00
100.
0095
.00
100.
0010
0.00
80.0
0%
Cei
ling
16.0
59.7
02.
72.
738
.770
.56.
7%
Floo
r0.
716
.10.
71.
30
1.3
13.4
0V
alid
NN
=15
0N
=14
9N
=15
0N
=15
0N
=15
0N
=15
0N
=14
9N
=15
0A
ges
60-6
9Fe
mal
e(N
:64)
PFRE
PBP
GHVT
SFRE
EM
HM
ean
64.1
468
.75
61.5
954
.27
60.0
079
.69
76.0
470
.88
25th
Perc
entil
e50
.00
25.0
041
.00
34.1
745
.00
62.5
041
.67
53.0
050
thPe
rcen
tile
(Med
ian)
65.0
010
0.00
62.0
059
.00
60.0
087
.50
100.
0072
.00
75th
Perc
entil
e83
.75
100.
0074
.00
73.5
075
.42
100.
0010
0.00
87.0
0St
dD
evia
tion
22.6
740
.82
18.8
624
.32
17.3
721
.54
38.7
018
.41
Ran
ge85
.00
100.
0068
.00
92.0
070
.00
100.
0010
0.00
76.0
0%
Cei
ling
4.7
57.8
03.
20
32.8
68.8
9.4
%Fl
oor
017
.20
00
1.6
15.6
0V
alid
NN
=64
N=
64N
=64
N=6
3N
=64
N=6
4N
=64
N=6
4A
ges
70&
over
Fem
ale
(N:1
8)PF
REP
BPG
VTSF
REE
MH
~M
ean
60.8
373
.61
63.7
851
.50
55.5
672
.22
77.7
869
.56
CD25
thPe
rcen
tile
36.2
531
.25
51.9
235
.83
41.2
540
.62
69.4
456
.50
Q..
.
<-
50th
Perc
entil
e(M
edia
n)70
.00
100.
0062
.00
46.0
051
.25
84.3
710
0.00
66.0
0~ 0
75th
Perc
entil
e94
.17
100.
0087
.590
.00
87.5
010
0.00
100.
0088
.00
0- '<St
dD
evia
tion
30.9
839
.73
20.8
229
.89
24.4
332
.24
39.6
118
.82
'" 0' ~R
ange
100.
0010
0.00
68.0
097
.00
85.0
087
.50
100.
0060
.00
01%
Cei
ling
5.6
66.7
00
038
.972
.216
.7(X
l z%
Floo
r5.
611
.10
5.6
00
16.7
00 01
Val
idN
N=1
8N
=18
N=1
8N
=18
N=1
8N
=18
N=1
8N
=18
0 CD l"l
CD 3 cr ~ tv 0 0 w
~Ta
ble
V:N
orm
sfo
r.Mal
esby
Age
Gro
up.
CD a...
<-
Ages
18-2
9M
ale
(N=2
62)
PFRE
PBP
GH
VTSF
REE
MH
~ 0M
ean
92.4
083
.72
71.8
670
.18
67.8
283
.83
74.1
772
.75
0"
-<25
thPe
rcen
tile
V>
90.0
075
.00
62.0
057
.00
55.0
075
.00
33.3
360
.00
c' Q:50
thPe
rcen
tile
(Med
ian)
100.
0010
0.00
74.0
072
.00
70.0
087
.50
100.
0074
.00
01
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
85.0
00
0 zSt
dD
evia
tion
14.5
228
.98
17.9
017
.65
16.5
419
.08
38.3
617
.00
0 01
Ran
ge10
0.00
100.
0080
.00
100.
0080
.00
100.
0010
0.00
84.0
00 CD
%C
eilin
g53
.869
.00
3.8
2.7
42.4
64.1
7.6
«)
nc
CDQ
...3
%Fl
oor
0.4
5.4
00.
40
0.4
15.6
0cr
..z~
Val
idN
N=
262
N=2
61N
=26
2N
=26
2N
=26
2N
=26
2N
=26
2N
=26
2g..
,'"0
Ages
30-3
9M
ale
(N=4
21)
PFRE
PBP
GH
VTSF
REE
MH
c0
CD'
wg..
,M
ean
91.4
584
.58
72.2
969
.29
69.5
684
.23
82.5
476
.28
3'-
25th
Perc
entil
e90
.00
75.0
062
.00
55.6
755
.00
75.0
010
0.00
64.0
0CD ~
50th
Perc
entil
e(M
edia
n)95
.00
100.
0074
.00
72.0
070
.00
87.5
010
0.00
76.0
0a c
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
88.3
3'7
)l c'St
dD
evia
tion
13.1
430
.34
16.8
018
.83
17.0
518
.69
33.9
416
.02
:> G)
Ran
ge90
.00
100.
0070
.00
90.0
085
.00
87.5
010
0.00
76.0
0CD :> CD
%C
eilin
g46
.874
.80
3.1
6.2
44.0
75.8
9.0
a%
Floo
r0
6.9
0.2
00
011
.00
-u 0 "'C
Val
idN
N=4
21N
=42
0N
=42
0N
=421
N=
420
N=
420
N=
418
N=
420
c cAg
es40
-49
Mal
e(N
=429
)PF
REP
BPG
HVT
SFRE
EM
Hg. ~
Mea
n88
.59
86.2
472
.11
69.6
170
.63
86.6
684
.39
78.2
6;:>
CCD en
25th
Perc
entil
e85
.00
75.0
062
.00
57.0
060
.00
75.0
010
0.00
68.0
0c vr
50th
Perc
entil
e(M
edia
n)95
.00
100.
0074
.00
72.0
070
.00
87.5
010
0.00
80.0
0a
75th
Perc
entil
e10
0.00
100.
0090
.00
85.0
080
.00
100.
0010
0.00
89.3
33 a
Std
Dev
iatio
n15
.39
27.7
216
.11
18.6
317
.117
.02
32.3
915
.26
-u 0 en
Ran
ge10
0.00
100.
0068
.00
95.0
080
.00
87.5
010
0.00
72.0
0a- U
>
%C
eilin
g33
.8·7
4.7
02.
37.
049
.478
.511
.7c
%Fl
oor
0.5
4.7
00
00
9.4
0~
Val
idN
cN
=42
9N
=42
7N
=42
9N
=42
9N
=42
9N
=42
9N
=42
7N
=42
9en S
'co 3'
-CD U
>'-
I7
'0
ww
0-
'IA
ges
50-5
9M
ale
(N=2
60)
PFRE
PBP
GH
VTSF
REE
MH
00
;;0
.I>-
Mea
n83
.69
82.1
270
.43
64.5
568
.46
85.7
783
.66
76.0
9C5
25th
Per
cent
ile75
.00
75.0
062
.00
52
.00
55.0
075
.00
83.3
361
.00
Z ~50
thP
erce
ntile
(Med
ian)
90.0
010
0.00
74.0
064
.50
70.0
087
.50
100.
0080
.00
» ;;0
75th
Per
cent
ile98
.75
100.
0090
.00
82.0
080
.00
100.
0010
0.00
92.0
0-l n
Std
Dev
iatio
n17
.90
32
.17
17.2
221
.51
17.4
018
.07
32.0
617
.60
.----
rn
Ran
ge85
.00
100.
0070
.00
100.
0085
.00
75.0
010
0.00
68.0
0
%C
eilin
g25
.071
.90
1.2
5.4
49.2
75.3
11.5
%Fl
oor
07
.70
1.2
00
9.3
0
Val
idN
N=
260
N=
26
0N
=26
0N
=2
60
N=
26
0N
=26
0N
=25
9N
=2
60
Age
s60
-69
Mal
e(N
=128
)PF
REP
BPG
HVT
SFRE
EM
HM
ean
75.2
066
.99
64.2
458
.36
62
.97
80
.47
79.4
375
.31
25th
Per
cent
ile65
.00
25.0
051
.00
45
.00
50.0
065
.63
66.6
764
.00
50th
Per
cent
ile(M
edia
n)80
.00
100.
0062
.00
61.0
060
.00
87.5
010
0.00
76
.67
75th
Per
cent
ile93
.75
100.
0084
.00
72.0
075
.00
100.
0010
0.00
88.0
0
Std
Dev
iatio
n21
.24
41.6
121
.12
20.4
118
.02
22.0
036
.26
17
.27
Ran
ge90
.00
100.
0090
.00
92.0
100.
0087
.50
100.
0084
.00
%C
eilin
g11
.755
.50
03.
138
.372
.78.
6
%Fl
oor
019
.52.
30
0.8
012
.50
Val
idN
N=
128
N=
128
N=
128
N=
128
N=
128
N=
128
N=
128
N=
128
Age
s70
&ov
erM
ale
(N=4
3)PF
REP
BPG
HV
ISF
REE
MH
Mea
n65
.81
65.1
262
.44
52.2
856
.51
72.9
771
.32
72.1
9~
25th
Per
cent
ile4
9.1
7.0
048
.75
35.0
039
.17
62.5
027
.78
54.0
0(!
) a...
'-
50th
Per
cent
ile(M
edia
n)70
.00
100.
0062
.00
50.0
053
.33
75.0
010
0.00
72.0
0~ c
75th
Per
cent
ile89
.17
100.
0079
.00
73.2
57
9.1
710
0.00
100.
0094
.00
Q '<St
dD
evia
tion
24~83
45.0
319
.52
24.4
125
.60
28.3
442
.15
21.0
9en 0
' ~R
ange
100.
0010
0.00
90.0
097
.00
100.
0010
0.00
100.
0084
.00
CJ1
%C
eilin
g4
.758
.10
09.
334
.965
.116
.3ex
> Z%
Floo
r2.
327
.92.
34
.72.
32.
320
.90
0V
alid
NCJ
1N
=4
3N
=43
N=
43N
=43
N=
43N
=43
N=
43N
=4
30 (!
)() (!) 3 lJ ~ N 0 0 w
s:Ta
ble
VI:
Nor
ms
for
Ethn
icG
roup
s,G
ener
alM
alay
sian
Popu
latio
n,Fe
mal
es&
Mal
esC
ombi
ned
CD a...
'-
-s:
Mal
ay(N
=20
19)
PFRE
PBP
GH
VTSF
REE
MH
c cM
ean
87.0
081
.81
69.5
867
.09
67.8
184
.44
79.5
875
.83
-< V>
25th
Perc
entil
eo'
80.0
075
.00
62.0
052
.00
55.0
075
.00
66.6
764
.00
~50
thPe
rcen
tile
(Med
ian)
95.0
010
0.00
72.0
067
.00
70.0
087
.50
100.
0076
.00
01
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
88.0
0(X
l zSt
dD
evia
tion
16.9
832
.33
17.3
219
.96
17.5
918
.98
35.8
617
.03
0 01
Ran
ge10
0.00
100.
0090
.00
100.
0010
0.00
100.
0010
0.00
100.
000 CD
%C
eilin
g33
.870
.40
2.1
4.5
44.8
71.9
10.5
0n
cCD
Q...
3%
Floo
r0.
18.
20.
30.
20.
10.
112
.80.
0..z
lJ ~V
alid
NN
=20
19N
=20
14N
=20
18N
=20
18N
=20
19N
=20
19N
=20
14N
=20
19Q
...
'"0c:
0Ch
ines
e(N
=505
)PF
REP
BPG
HVT
SFRE
EM
HCD
'w
Q...
Mea
n85
.52
85.6
974
.52
67.3
265
.37
82.4
382
.67
73.2
6s=
-25
thPe
rcen
tile
55.0
010
0.00
64.0
0CD
80.0
010
0.00
62.0
052
.00
75.0
0~
50th
Perc
entil
e(M
edia
n)90
.00
100.
0080
.00
67.5
065
.00
87.5
010
0.00
73.3
3c C
-
75th
Perc
entil
e10
0.00
100.
0090
.00
82.0
080
.00
100.
0010
0.00
88.0
0~ o'
Std
Dev
iatio
n::
l
17.5
029
.87
17.2
719
.28
17.3
619
.53
33.5
416
.89
G)
85.0
010
0.00
90.0
010
0.00
95.0
010
0.00
100.
0092
.00
CDR
ange
::l
CD
%C
eilin
g28
.577
.20
3.6
3.2
39.6
75.9
6.1
a -c%
Floo
r0
6.3
0.2
0.2
00.
210
.20
0 ""U c
Val
idN
N=5
05N
=50
5N
=50
5N
=50
4N
=50
5N
=50
5N
=50
2N
=50
5C
-
Indi
an(N
=16
5)RE
EM
Hg-
PFRE
PBP
GH
VTSF
::l
Mea
n78
.76
79.2
466
.00
65.0
261
.55
79.7
374
.80
67.8
5;>
0CD '"c
25th
Perc
entil
e68
.33
75.0
052
.00
52.0
050
.00
65.6
233
.33
56.0
0ur
50th
Perc
entil
e(M
edia
n)85
.00
100.
0068
.00
65.0
065
.00
87.5
010
0.00
69.3
3a 3
75th
Perc
entil
e95
.00
100.
0084
.00
82.0
075
.00
100.
0010
0.00
80.0
0c "
Std
Dev
iatio
n20
.99
31.0
719
.29
19.7
217
.75
20.8
336
.20
17.9
60 ff
Ran
ge10
0.00
100.
0090
.00
100.
0090
.00
87.5
010
0.00
84.0
0- en c
%C
eilin
g21
.858
.80
0.6
1.8
32.3
62.2
3.6
~%
Floo
r0.
67.
90.
61.
20
011
.60
cV
alid
NN
=16
5'"
N=1
65N
=16
5N
=16
5N
=16
5N
=16
5N
=16
4N
=16
4::
;'cc s=
-CD en
"--
n0
W0
10
-
'IO
ther
Bum
iput
era
(N=3
54)
00
PFRE
PBP
GH
VTSF
REE
MH
0-
;:>
0
Mea
n84
.04
78.3
566
.98
64.5
765
.20
83.5
374
.83
73.0
1Q z
25th
Perc
entil
e75
.00
70.8
352
.00
52.0
050
.00
75.0
033
.33
60.0
0» .-
50th
Perc
entil
e(M
edia
n)95
.00
100.
0062
.00
65.3
365
.00
87.5
010
0.00
76.0
0» ;:
>0
75th
Perc
entil
e10
0.00
-l
100.
0084
.00
82.0
080
.00
100.
0010
0.00
88.0
0n .-
Std
Dev
iatio
n20
.72
34.6
117
.63
21.2
018
.18
19.2
538
.58
17.5
8rn
Ran
ge10
0.00
100.
0080
.00
100.
0090
.00
87.5
010
0.00
80.0
0
%C
eilin
g33
.965
.00
3.1
3.7
44.2
66.1
9.1
%Fl
oor
0.3
11.1
00.
80
016
.00
Val
idN
N=3
54N
=351
N=3
53N
=354
N=3
53N
=353
N=3
51N
=353
Oth
ers
(N=2
9)PF
REP
BPG
HVT
SFRE
EM
H
Mea
n87
.24
93.9
775
.69
68.3
870
.52
81.9
073
.56
76.2
825
thPe
rcen
tile
82.5
010
0.00
62.0
058
.50
57.5
070
.83
44.4
468
.00
50th
Perc
entil
e(M
edia
n)95
.00
100.
0074
.00
70.3
375
.00
91.6
710
0.00
77.3
375
thPe
rcen
tile
100.
0010
0.00
90.0
087
.00
80.0
010
0.00
100.
0084
.00
Std
Dev
iatio
n21
.11
19.6
614
.68
20.0
214
.10
23.7
639
.22
12.6
5
Ran
ge10
0.00
100.
0048
.00
85.0
060
.00
100.
0010
0.00
48.0
0
%C
eilin
g34
.586
.20
3.4
3.4
48.3
62.1
3.4
%Fl
oor
3.4
3.4
00
03.
417
.20
Val
idN
N=
29N
=29
N=2
9N
=29
N=2
9N
=29
N=
29N
=29
~ (I) a..
'- ~ Q Q '< V> c' Q: 01 co Z 0 01 0 (I)
() (I) 3 0-
~ N 0 0 w
Quality of Life of the Malaysian General Population: Results from a Postal Survey Using the SF-36
120
100
80 .PF
_REP60
.BP
40 .GH
20I/Ilvr
@0 00 IIJREEIf)
-..I0
.MHd -20Male Female
Gender
Fig. 1: Box plot for eight subscales of SF·36, by sex
SF
.PF
_REP
.SP
.GH++-t-l--I
IBVT20 ,j---=:.J-'-+-----'L.f-!----...L+..J.....;.-----'-++---''----+++t-H----/
80
40 ,j---H-+-
60 +---+
120
100
~~ 0 -1- ...1..- ...1..-__--'-1.-'-__--/ • REE
-..I
g -20 _MH18 - 29 30 - 39 40 -49 50 - 59 60 +
Age
Fig. 2: Box plot for eight subscales of SF·36, by age
Med J Malaysia Vol 58 No 5 December 2003 707
ORIGINAL ARTICLE
120
100
80
60
40
20
(1)....0
0U(f)
...J0a -20
Malay Chinese
Ethnic Group
Indian OtherBumi Others
.PF
_REP
.BP
_GH
B1VTSF
REE
_MH
Fig. 3: Box plot for eight subscales of SF·36, by ethnicity
Comparison of QOL Mean Scores for General Population
.---~-----------------------------------100
90
~ 80a'-'
Cf)
c:m(])
:2 70
60
50
I________________ ~ ~ __ ~ J
-L p- I
J-"''"'ni{:i=~-mmm-m'----mmnn;:.;h:l''n~nFn---nnm-
- P.... I- F_- c.-c - - - -- - - - --- - - ---- - --- t-=-~ ------ to---------------------------------------1-- -- --..-- -.
I---------------1-1-- . I- '_____________ -- --- __ ---_--- -- t:: --_.."..,.1:'::_ -- -- __ -- -.1
I-------.---------------r-----------------------
q:/-z--,<Iluq:q:::>q:O:2i:i:±r.:2:2:2:2 __
rUpperhmllMeanLowerl1mit
q:/-z--,CIlOq:q:::>q:O:2;wwwwWWWW0::0::0::0::
QOL Health Dimension
Fig. 4: Comparison of QOL Mean Scores for General Population for some countries
708 Med JMalaysia Vol 58 No 5 December 2003
Qualify of Life of the Malaysian General Population: Results from a Postal Survey Using the SF-36
Discussion
Health of the population in relation to functionalhealth, well-being, and relative burden of disease areessential information to evaluate the effectiveness ofany health care service across diverse populations. Thedearth of such information can be attributed todifficulties in finding measurement tools that is bothappropriate and practical to apply. We have used theSF-36 because of its well-established evidence thatsuggests the instrument to be sensitive to changes inthe health of the general population6,7.
Normative data make it possible to interpret the scalescore for an individual respondent or the average scorefor a group of respondents in comparison to thedistribution of scores for other individuals in themorning sample. For normative data to be valid, theymust be based on a well-defined and representativesample of the population4
• Though effort had beenmade in the sampling design to increaserepresentativeness, with a response rate of only 30.6%and significant age-sex composition and ethnicdifferences between the sample and the population,readers are cautioned on the possible influence ofresponse bias.
The means and standard deviations are presented toenable comparison of individuals or specific groups ofpopulations' scores with the Malaysian average.Caution is advised in interpreting and using the resultsdue to poor response rate. Furthermore, somesubgroups have smaller sample sizes and estimates maynot be stable. This is especially so for the Others ethnicgroup, and those 70 and above. Care also needs to beexercised when interpreting the results for SF and BPdue to the errors in the questionnaire.
In addition, the scope of these results is limited toMalaysian adults aged 18 years and above, and thoseliterate in either Bahasa Malaysia or English Language.Response bias may also affect the means, given thepoor response rate of only 30.6%, though reportedrates for postal surveys have been reported to rangefrom a low of 24% to a high of 92%!8.
In general, the mean scores for all scales were above65.0 QOL units. On a scale of 0-100, the higher scoresmay be interpreted as having achieved substantialquality in their life with the population perceiving to bein better health physically than mentally. Thevariability in scores by age and sex underscores the
Med J Malaysia Vol 58 No 5 December 2003
need to use the appropriate age- or sex-specificnormative data whenever possible.
Our findings show that there are significant genderdifferences within the Malaysian population, with menscoring higher on all domains. These results areconsistent with those reported for the United States,United Kingdom and Canada12
•
Our study did not show any significant differencesbetween ethnic groups. In addition, lower scores werereported for MH among the younger population,perhaps reflecting higher expectations. Further indepth qualitative studies are required to attemptexplaining these phenomena.
Overall, means for the Malaysian population generallydiffer from the population norms for USA\ Canada!2and the ACT, Australia13 with the exception of REE,which is noted to be a less sensitive scale4
• Thedifferences between countries could be due tomethodological dissimilarity instead of reflecting truedifferences. Possibly, there may be differences in agesex composition of the general population between thecountries, differences in prevalence of co morbidity,and problems with cross-population comparabilitybetween the countries. The norms were not adjusted toa standard population composition, as nOrms fromother countries were not presented in that manner andhence comparison would be difficult. Problems withcross population comparability include inconsistentreporting between actual and self-reported health statesand differences in end-points and cut-points on thereference scale19 . Biases in self-report of health statusin non-fatal outcomes had been reported to affectcomparison across populations!9. Here, the question iswhether, for example, the sexes, differ in theirwillingness to endorse lower end (poorer functioning)items, or whether it actually reflects poorer Quality ofLife for women. We assumed that the distribution ofQOL is equivalent for both sexes, and it is the reportingthat differs. Hence, we have reported separate norms20
•
Most subscales of SF-36 are sensitive for the Malaysiangeneral population, with the exception of REP and REE.Sensitivity of the subseaIe PF increases with age. Theceiling effects noted were most conspicuous for theREP and REE. This was not surprising as these twoscales of the SF-36 are relatively coarse role disabilityscales2!. Both measures have only four and five levelsrespectively across a restricted range and, therefore,usually have the most problems with ceiling and floor
709
ORIGINAL ARTICLE
effects. Knowing the fact that ceiling and floor effectsdoes exist is important since the ability of an instrumentto detect change is constrained by the percentage ofrespondent at either end of the effects.
The differences seen in the SF-36 scores across age,gender and countries confirm that these Malaysiannorms are essential. It is useful for comparison withdiseased state values and means in studies in Malaysia.It may be used as a measure to reflect the "shortfall" inquality of life for a diseased or handicapped state whencompared to the general Malaysian population. Thisdistribution can also be used as a baseline forcomparison in future surveys looking at quality of lifeof the general population, especially afterinterventional programmes of public health nature.
II
1. Guyatt GH and Cook DJ. Health status, quality of life andthe individual. Journal of American Medical Association.1994; 272: 630-31.
2. Bowling A Measuring Health. A Review Of Quality OfLife Measurement Scales. Buckingham: Open UniversityPress, 1991.
3. Wilson lB, Cleary PD. Linking clinical variables withhealth-related quality of life. JAMA 1995; 1995: 59-65.
4. Ware JE Jr, Kristin KS, Kosinski M, et al. SF-36 HealthSurvey manual and Interpretation Guide. Boston,Massachusetts: Nimrod Press, 1993.
5. Shiely JC, Boyliss MS, Keller SD, et al. SF-36 HealthSurvey Annotated Bibliography: First Edition (1988 1995). Boston, Massachusetts: The Health Institute, NewEngland Medical Centre, 1996.
6. Hemingway H, Stafford M, Stansfeld S, Shipley M &Marmot M. Is the SF-36 a valid measure of change inpopulation health? Results from the Whitehall II study.BMJ 1997; 315: 1273-79.
710
Acknowledgements
This study received a grant from the IntensifiedResearch Priority Areas Grant, Government of Malaysia(IRPA Code:06-05-05-0177). The authors would like toexpress their gratitude to the Department of StatisticsMalaysia including Hjh. Normah Mohd Aris, Head,Communications And Operations Division, Kate Van KitYee, Rosilawati, Rafliza and Kho Sian Eng and all thestate statisticians for their invaluable assistance inplanning and implementation of the data collectionphase of this survey. Their assistance, support andagreement to allow this survey to piggyback on thethird Round of the Labour Force Survey is very muchappreciated and valued. The researchers are alsograteful to Prof Wan Manan and team, UniversityScience Malaysia (USM) for their assistance and forproviding their developmental version of the translatedMalay version of SF-36. All their support andcooperation given is very much appreciated.
IHii
7. Watson EK, Firman DW, Baade PD, Ring I. Telephoneadministration of the SF-36 Health Survey: validationstudies and population norms for adults in Queensland.Aust NZ J Public Health 1996; 306: 1437-40. [MEDLINEI
8. Ware JE Jr, Sherbourne CD. The MOS 36-item Short-FormHealth Survey (SF-36): I. Conceptual framework and itemselection. Med Care 1992; 30(6): 473-83.
9. McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-itemShort-Form Health Survey (SF-36):II. Psychometric andClinical Tests of Validity in Measuring Physical and MentalHealth Constructs. Med Care 1993; 31(2): 247-63.
10. McHorney CA, Ware JE Jr, Lu JFR & Sherbourne CD. TheMOS 36-item Short-Form Health Survey (SF-36): III. Testsof data quality, scaling assumptions, and reliability acrossdiverse patient groups. Med Care 1994; 32(1): 40-66.
11. Ware JE Jr, Kosinski M, Bayliss MS, McHorney CA, RogersWH, Raczek A. Comparison of Methods for the Scoringand Statistical Analysis of SF-36 health profile andsummary measures: summary results from the MedicalOutcomes Study. Med Care 1995; 33(4 Supp!): AS264AS279.
Med J Malaysia Vol 58 No 5 December 2003
Quality of life of the Malaysian General Population: Results from a Postal Survey Using the SF-36
12. Wilma M Hopman, Tanveer Towheed, TassosAnastassiades, Alan Tenenhouse, et al. CanadianNormative Data for the SF-36 Health Survey. CanadianMedical Association J 2000; 163(3): 265-71.
13. Gannon D, Gordon C, Egloff B & Shadbolt B. HealthRelated Quality of Life in the ACT: 1994-1995. TheBaseline. Australia: ACT Department of Health &
Community Care, Population Health Group,Epidemiology Unit, 1997.
14. Jenkinson C, Stewart-Brown S, Peterson S, Paice C.
Assessment of the SF-36 version 2 in the United Kingdom.J Epidemiol Community Health 1999; 53: 46-50.[MEDLINE].
15. Sararaks S, Rugayah B, Azman AB, Karuthan C & Low LL.Quality of life: How do Malaysian asthmatics fare? . MedJ Malaysia 2001; 56(3): 350-58.
16. Population and Housing Census of Malaysia 2000:Population Distribution and Basic DemographicCharacteristics. Malaysia: Department of Statistics, July2001.
Med J Malaysia Vol 58 No 5 December 2003
17. Ware JE Jr. SF-36 Health Survey Manual and InterpretationGuide. Boston: The Health Institute, New EnglandMedical Centre, 1993.
18. Basic Survey Design & Implementation. Module 4 Health surveys. Workshop on Evidence for Health Policy:Burden of Disease, Cost-effectiveness, Health SystemsPerformance and Health Surveys, Dikhololo, SouthAfrica. World Health Organization & Burden of DiseaseUnit, Harvard Center for Population and DevelopmentStudies, Oct 2001.
19. Sadana R, Mathers CD, Lopez AD, Murray CJL & Iburg K.
Comparative analysis of more than 50 household surveyson health Status. World Health Organization GPEDiscussion Paper Series No. 15 EIP/GPE/EBD.
20. Streiner DL & Norman GR. Health Measurement Scales. Apractical guide to their development and use. New York:Oxford University Press, 1995.
21. Ware JE Jr & Gandek B. Overview of the SF-36 HealthSurvey and the Intenational Quality of Life Assessment(IQOLA) Project. J Clin Epidemiol 1998; 51(11): 903-12.
711
