quality of life in dialysis: a malaysian perspective

Quality of life in dialysis:A Malaysian perspective

Wen J. LIU,1 Ramli MUSA,2 Thian F. CHEW,3 Christopher T. S. LIM,4 Zaki MORAD,5

Adam BUJANG6

1Department of Medicine, Sultanah Aminah Hospital, Johor Bahru, Johor, Malaysia; 2Department ofPsychiatry, International Islamic University, Kuantan, Pahang, Malaysia; 3Seremban KPJ SpecialistHospital, Seremban, Negeri Semibilan, Malaysia; 4University of Putra Malaysia, Serdang, Malaysia;

5Ampang Puteri Hospital, Kuala Lumpur, Malaysia; 6Biostatistics Department, Clinical Research Centre,Kuala Lumpur, Malaysia

AbstractThere is a growing interest to use quality of life as one of the dialysis outcome measurement. Basedon the Malaysian National Renal Registry data on 15 participating sites, 1569 adult subjects whowere alive at December 31, 2012, aged 18 years old and above were screened. Demographic andmedical data of 1332 eligible subjects were collected during the administration of the short formof World Health Organization Quality of Life questionnaire (WHOQOL-BREF) in Malay, English, andChinese language, respectively. The primary objective is to evaluate the quality of life amongdialysis patients using WHOQOL-BREF. The secondary objective is to examine significant factorsthat affect quality of life score. Mean (SD) transformed quality of life scores were 56.2 (15.8), 59.8(16.8), 58.2 (18.5), 59.5 (14.6), 61.0 (18.5) for (1) physical, (2) psychological, (3) social relations,(4) environment domains, and (5) combined overall quality of life and general health, respectively.Peritoneal dialysis group scored significantly higher than hemodialysis group in the mean combinedoverall quality of life and general health score (63.0 vs. 60.0, P < 0.001). Independent factors thatwere associated significantly with quality of life score in different domains include gender, bodymass index, religion, education, marital status, occupation, income, mode of dialysis, hemoglobin,diabetes mellitus, coronary heart disease, cerebral vascular accident and leg amputation. Subjectson peritoneal dialysis modality achieved higher combined overall quality of life and general healthscore than those on hemodialysis. Religion and cerebral vascular accident were significantlyassociated with all domains and combined overall quality of life and general health.

Key words: Continuous ambulatory peritoneal dialysis, dialysis, hemodialysis, short form ofWorld Health Organization Quality of Life, quality of life

Correspondence to: W. J Liu, MRCP, Department of Medicine, Sultanah Aminah Hospital, Johor Bahru 80100, Johor,Malaysia. E-mail: [email protected] authors declare no conflict of interest.This study was registered with the National Medical Research Register (NMRR registration no. 11-827-10135) and approvedby the Medical Research Ethics Committee of Malaysia.This study is a collaborative effort between National Renal Registry, Malaysia and the Psychiatry Department of InternationalIslamic University of Malaysia. Funding comes from the research grant of the Ministry of Science, Technology and Innovationof Malaysia.

Hemodialysis International 2014; 18:495–506

© 2013 International Society for HemodialysisDOI:10.1111/hdi.12108

495

INTRODUCTION

With the expansion of dialysis provision in this country,dialysis therapy has prolonged survival of a substantialnumber of those with end-stage renal disease (ESRD).Hemodialysis (HD) providers in Malaysia can be dividedinto three sectors: the government, the nongovernmentorganizations (nonprofit-making, charitable groups), andthe private centers. Ninety-nine percent of peritonealdialysis (PD) is provided by the government sector at anominal cost. The cost of dialysis therapy can be prohibi-tively expensive especially for those who are of poor socio-economic status. Only 22% of the 25,688 dialysis patientsin 2011 are capable of self-funding.1 The remaining 78%obtain financial assistance either from the government,social security organization, religious, or charity groups.There is a growing interest to use quality of life (QOL) asone of the outcome measurement in dialysis cohort. Thisrationale assumes QOL has a close relationship with mor-bidity and mortality.2,3 The use of Spitzer’s Quality of LifeIndex Score by the Malaysian National Renal Registry1

(NRR) for annual reporting has its deficiencies as thisinstrument was hampered by poor discriminatory ability.4

The World Health Organization (WHO) defines QOL as“individuals’ perceptions of their position in life in thecontext of the culture and value systems in which theylive, and in relation to their goals, expectations, standards,and concerns.”5 In 1991, the WHO initiated a cross-cultural project to develop a 100 items QOL questionnaire(WHOQOL) for generic use.6 Shortly, a short form wasdeveloped and named WHOQOL-BREF.7 It has a total of26 items regarded as important by 15 field centers ofdifferent countries in assessing QOL. With similar psycho-metric properties, the WHOQOL-BREF has been shownto be a valid and reliable alternative to the assessment ofdomain profiles using the WHOQOL-100.6 It is also areliable and valid QOL assessment tool.8

We explored the use of WHOQOL-BREF in dialysispatients. This study aims to measure QOL score of dialysispatients in four domains (physical, psychological, socialrelationships, and environment) and overall QOL andgeneral health according to the WHOQOL-BREF.

MATERIALS AND METHODS

This is an investigator-initiated, multicenter study con-ducted from May to October 2012 at 15 centers thatprovide HD and/or PD. These centers comprised of theMinistry of Health hospitals (n = 5), Government Univer-sity hospitals (n = 5), nongovernment organizations(n = 3), and private hospitals (n = 2). The primary objec-

tive is to evaluate the QOL among dialysis patients usingWHOQOL-BREF. The secondary objective is to examinestatistically significant factors that affect QOL score. Theauthors invited 15 dialysis centers to recruit a target of atleast 1000 subjects that represented the three HD sectors.Selection of participating sites was done by taking intoconsideration of various factors such as hospital setting(private vs. government hospitals), racial distribution, andavailability of local clinical research staff at various dialysiscenters. Based on the database of National Renal Registry asof December 31, 2012, selection of 1569 potential patientswere universal/all inclusive. Inclusion criteria are subjectsof 18 years old and above on dialysis for at least 3 months,ability to communicate in Malay, Chinese, or English,ability to give informed consent. Those who are sufferingfrom dementia/delirium or have emergency hospitalizationfor more than 7 days within the past month are excluded.Those eligible subjects who consented are administered(self or assisted by trained personnel) WHOQOL-BREFquestionnaire in either Malay,9,10 English11, or Chinese12

language. Demographic data (mode of dialysis, age, gender,religion, education, marital status, occupation, income)and medical information (body mass index, duration ondialysis, hemoglobin, diabetes mellitus [DM], hyperten-sion, coronary heart disease [CHD], cerebral vascular acci-dent [CVA], leg amputation, cancer, visual impairment,history of psychiatric illness) were collected from medicalrecords. During the interim period of this study,WHOQOL-BREF in Chinese and English were tested andfound to be valid and reliable. All research assistants weretrained in administering the questionnaire during the studyinitiation meeting conducted on May 15, 2012.

This study was registered with the National MedicalResearch Register (NMRR registration no. 11-827-10135)and approved by the Medical Research Ethics Committeeof Malaysia.

Statistical analysis

The WHOQOL-BREF raw scores of each domain wereconverted to transformed scores (0–100) according toprotocol.13 Higher scores reflect better QOL. Fourdomains of physical health, psychological, social relations,environment as well as subjective overall QOL and generalhealth were analyzed using the general linear model forunivariate and multivariate analyses. All significant factorsby multivariate analyses were tested by least significantdifference multiple comparison test. Crude and adjusted Pvalues were stated to indicate significance of associationbetween factors and outcome. All factors that showedstatistical significance in univariate analyses were included

Liu et al.


in multivariate analyses. Multi-colinearity was checkedbefore the conduct of multivariate analysis in order toconfirm that all factors were independent of each other.For example, Malay race (ethnicity) and Islamic faith (reli-gion) were highly dependent on each other. Therefore,only one variable was selected for the multivariate analysisalthough both variables were significant. Marginal meanof QOL scores with 95% confidence interval werereported for factors of categorical value while correlationcoefficient with r-squared were reported for factors ofnumerical value. All analyses were carried out using SPSS(IBM SPSS version 20.0, IBM Corp., Armonk, NY, USA).Missing values were managed according to WHOQOL-BREF scoring instruction.13

RESULTS

Screening was done among 1569 subjects who were aliveat December 31, 2012 from the 15 selected centers. A total

of 1332 eligible subjects were enrolled in the study afterobtaining consent. Two hundred thirty-seven cases werenot included in this study with the following reasons:recent emergency hospitalization (n = 52), organic brainsyndrome (n = 18), no permanent vascular access (n = 4),refusal (n = 84), change modality of dialysis, or death(n = 79). Baseline characteristics are listed in Table 1.Majority of the subjects are Muslims. Seventy-four percenthad a monthly family income of <RM3000 (USD 1000).DM and hypertension are the two commonest comorbidconditions found in 46.5% and 83.9% of subjects,respectively.

The mean (standard deviation) transformed scores (outof 100) of the respective four domains and combinedoverall QOL and general health of WHOQOL-BREFranged from 56.2 (15.8) to 61.0 (18.5) (Table 2). PDgroup had a statistically higher overall QOL score thanHD group. (P < 0.001) There were no statistical differ-ences between HD and PD groups in the QOL score of

Table 1 Baseline characteristics of subjects according to mode of dialysis

HD (793) PD (539) All (1332)

n % n % n %

Agea 55.5 15.3 52.8 15.4 54.4 15.4Male 439 55.4 240 44.5 679 51.0BMI

1. Underweight 92 11.7 35 6.5 127 9.62. Normal 392 50.1 254 47.5 646 49.03. Overweight 212 27.1 168 31.4 380 28.84. Obese 87 11.1 78 14.6 165 12.5

Race1. Malay 357 45.0 318 59.0 675 50.72. Chinese 331 41.7 160 29.7 491 36.93. Indian 95 12.0 54 10.0 149 11.24. Others 10 1.3 7 1.3 17 1.3

Religion1. Muslim 363 45.9 320 59.7 683 51.52. Buddhist 284 35.9 141 26.3 425 32.13. Hindu 81 10.3 48 9.0 129 9.74. Christian 49 6.2 18 3.4 67 5.15. Others 13 1.6 9 1.7 22 1.7

Education level1. Nil 79 10.3 24 4.5 103 7.92. Primary 189 24.5 158 29.4 347 26.53. Secondary 350 45.5 275 51.2 625 47.84. Tertiary 152 19.7 80 14.9 232 17.8

Marital status1. Single 138 17.4 96 17.9 234 17.62. Married 570 72.1 401 74.7 971 73.13. Widow/widower 74 9.4 33 6.1 107 8.14. Divorced 9 1.1 7 1.3 16 1.2

Quality of life in dialysis

Hemodialysis International 2014; 18:495–506 497

four domains (Table 2). Religion and CVA both were sig-nificantly associated with QOL scores in all domains(Table 3).

Significant factors associated withphysical domain (Table 3)

Normal (P < 0.001) and overweight (P = 0.004) patients,respectively, had significantly higher physical score com-pared with obese patients. Muslim subjects showed sig-nificantly higher physical score compared with Hindus(P = 0.040) and Christians (P = 0.024). General workerhad significantly higher physical score compared withhomemaker (P = 0.001), unemployed (P = 0.001), andretiree (P = 0.047). Those with the lowest income

Table 2 WHOQOL BREF transformed scores according todomains reported in hemodialysis (HD), peritoneal dialysis(PD) and all groups

WHOQOL-BREFdomains

Mean QOL score ± SD

HD PD All

Physical* 56.6 (16.1) 55.5 (15.5) 56.2 (15.8)Psychological* 59.6 (17.3) 60.2 (16.0) 59.8 (16.8)Social relations* 57.6 (18.4) 59.2 (18.6) 58.2 (18.5)Environment* 58.9 (14.9) 60.4 (14.1) 59.5 (14.6)QOL and general

health**60.0 (19.2) 63.0 (17.1) 61.0 (18.5)

QOL = quality of life.*P > 0.05 by Student’s t test between HD and PD.**P < 0.001 by Student’s t test between HD and PD.

Table 1 Continued

HD (793) PD (539) All (1332)

n % n % n %

Occupation1. Professional 48 6.1 29 5.4 77 5.82. Semi-professional 44 5.6 20 3.7 64 4.83. General worker 74 9.5 48 8.9 122 9.24. Laborer 12 1.5 9 1.7 21 1.65. Homemaker 107 13.7 186 34.6 293 22.26. Retired 158 20.2 135 25.1 293 22.27. Unemployed 340 43.4 111 20.6 451 34.1

Monthly family income1. RM0–RM999 201 26.4 186 34.9 387 29.92. RM1000–RM1999 209 27.4 171 32.1 380 29.33. RM2000–RM2999 124 16.3 67 12.6 191 14.74. > RM3000 228 29.9 109 20.5 337 26.0

Types of center1. Government 390 49.2 539 100.0 929 69.72. Nongovernment organization 254 32.0 0 0.0 254 19.13. Private 149 18.8 0 0.0 149 11.2

Duration of disease in yearsa 7.6 6.2 3.8 3.1 6.1 5.5Hemoglobina 10.7 1.5 10.4 1.7 10.6 1.6Comorbidities

DM 353 44.5 266 49.4 619 46.5Hypertension 643 81.1 474 87.9 1117 83.9CHD 111 14.0 93 17.3 204 15.3CVA 30 3.8 10 1.9 40 3.0Leg amputation 26 3.3 17 3.2 43 3.2Cancer 12 1.5 4 0.7 16 1.2Vision impairment 48 6.1 43 8.0 91 6.8

BMI = body mass index; CHD = coronary heart disease; CVA = cerebral vascular accident; DM = diabetes mellitus; HD = hemodialysis;PD = peritoneal dialysis.aReported as mean (SD).

Liu et al.


showed significantly poorer physical score compared withthose whose earnings RM1000–RM1999 (P = 0.001),RM2000–RM2999 (P = 0.023), and RM3000 andabove (P < 0.001). Patients with DM (P = 0.007), CHD(P = 0.033), CVA (P = 0.008), and leg amputation(P = 0.024) reported poorer physical score than thosewithout (Table 4).

Significant factors associated withpsychological domain (Table 3)

Male had higher psychological score with female(P = 0.043). Muslim had significant higher psychologicalscore compared with Buddhist (P < 0.001), Hindu(P = 0.004), and Christian (P = 0.001). Hindu subjectshad better psychological health score compared with Bud-dhist (P = 0.011). Patients with tertiary education levelshowed significantly higher psychological score comparedwith the uneducated (P = 0.014), primary (P = 0.026),and secondary (P = 0.002). Patients with DM (P = 0.049),CHD (P = 0.041), and CVA (P = 0.008) had significantlypoorer psychological score compared with those without.Although hemoglobin had significant association withpsychological score, the effect size was relatively low(correlation coefficient = 0.069; Table 5).

Significant factors associated with socialrelations domain (Table 3)

Muslim subjects had better social score compared withBuddhists (0.006) and Hindus (< 0.001). Married sub-

jects had better social score compared with singles(<0.001). General worker had better social score com-pared with semi-professional (P = 0.025) and unem-ployed (0.001). Homemaker had better social scorecompared with the unemployed (P = 0.030). Patients withCHD (P = 0.010), CVA (P = 0.011), and leg amputation(P = 0.048) had poorer social score than those without(Table 6).

Significant factors associated withenvironment domain (Table 3)

Muslim subjects had better environment score comparedwith Buddhists (P < 0.001), Hindus (P = 0.005), Chris-tians (<0.001), and others (0.011). Hindu subjectsachieved higher environmental score compared with Bud-dhist (P < 0.001) and Christian (P = 0.012). Patients withtertiary level education showed better environment scorecompared with those who are uneducated (P = 0.004),primary (P = 0.004), and secondary (P = 0.002). Patientswith CVA (P = 0.033) had poorer environment score com-pared with those without. Although hemoglobin had sig-nificant association toward environment, the effect isrelatively low (correlation coefficient = 0.069; Table 7).

Significant factors associated withcombined overall QOL and general healthdomain (Table 3)

Patients with normal BMI had better overall QOL scorecompared with those who are underweight (P = 0.045).

Table 3 Multivariate analysis: Summary of all statistically significant demographic and medical factors associated withWHOQOL-BREF domains

P value

Physical Psychological Social Environment Overall

Gender 0.043BMI 0.003 0.024Religion 0.043 <0.001 0.003 <0.001 <0.001Education level 0.014 0.002Marital status 0.002Occupation 0.026 0.010 0.021Family income 0.001Mode of dialysis 0.031Hemoglobin 0.018 0.006 0.006DM 0.007 0.0049CHD 0.033 0.041 0.010CVA 0.008 0.008 0.011 0.033 0.004Leg amputation 0.004 0.048

BMI = body mass index; CHD = coronary heart disease; CVA = cerebral vascular accident; DM = diabetes mellitus.



On the other hand, overweight patients had better overallQOL compared with underweight (P = 0.017) and obese(P = 0.024) patients. Muslim subjects achieved betteroverall QOL score compared with Buddhists (P < 0.001),Christians (P < 0.001), and others (P = 0.006). Hindu hadbetter overall QOL compared with Buddhists (P < 0.001),Christians (P = 0.007), and others (P = 0.022). Those onPD had better overall QOL compared with patients withHD (P = 0.031). Patients with CVA (P = 0.004) hadpoorer overall QOL score compared with those without.

Although hemoglobin was significantly associated withoverall QOL and general health, the effect is relatively lowwith a correlation coefficient of 0.060 (Table 8).

DISCUSSION

According to Cagney et al., of the 53 different QOL instru-ments used, 82% were generic and 18% disease-specific,with the Sickness Impact Profile and Kidney DiseaseQuestionnaire having been more thoroughly validated

Table 4 Standardized marginal mean of QOL and least significant difference (LSD) multiple test comparison of allsignificant factors that are associated with physical domain

Variables Marginal means (95% CI) LSD multiple test comparison (P value)

BMI status1. Normal 44.0 (38.3,49.7) Normal > obese (<0.001)

Overweight > obese (0.004)2. Underweight 42.1 (35.8,48.3)3. Overweight 43.2 (37.4,49.0)4. Obese 38.8 (32.6,45.0)

Religion1. Muslim 43.6 (37.9,49.2) Muslim > Hindu (0.040)

Muslim > Christian (0.024)2. Buddhist 41.7 (36.0,47.5)3. Hindu 40.3 (34.2,46.5)4. Christian 38.5 (31.8, 45.2)5. Others 45.8 (37.0,54.6)

Occupation1. Professional 43.2 (36.3,50.1) Gen. worker > homemaker (0.001)

Gen. worker > unemployed (0.001)Gen. worker > retired (0.047)

2. Semi-professional 41.5 (34.6,48.4)3. General worker 46.4 (40.1,52.6)4. Laborer 39.1 (30.1,48.1)5. Homemaker 40.5 (34.7,46.2)6. Retired 42.6 (36.9,48.4)7. Unemployed 40.7 (35.1,46.3)

Family income1. RM0–RM999 39.0 (33.1,45.0) RM0–RM999 < RM1000–RM1999 (0.001)

RM0–RM999 < RM2000–RM2999 (0.023)RM0–RM999 < RM3000 and above (<0.001)

2. RM1000–RM1999 42.9 (37.0,48.8)3. RM2000–RM2999 42.4 (36.3,48.5)4. > RM3000 43.7 (37.9,49.5)

DM1. Yes 40.6 (34.9,46.4) No > yes (0.007)2. No 43.3 (37.5,49.2)

CHD1. Yes 40.6 (34.7,46.6) No > yes (0.033)2. No 43.4 (37.6,49.1)

CVA1. Yes 38.5 (31.4,45.6) No > yes (0.008)2. No 45.5 (40.2,50.7)

Leg amputation1. Yes 39.0 (39.7,50.3) No > yes (0.024)2. No 45.0 (39.7,50.3)

CI = confidence Interval; CHD = coronary heart disease; CVA = cerebral vascular accident; DM = diabetes mellitus; > = better QOL score than.

Liu et al.


than others.14 Use of QOL instruments in the Malaysianscene has its unique considerations. Malaysia has threemajor ethnic groups: the Malays accounting for the major-ity, followed by the Chinese, and the Indians. We used thethree translations of WHOQOL-BREF questionnairesfrom the public domain to study QOL among dialysispopulation. Validation of the Malay translation ofWHOQOL-BREF has been performed in the Malaysianpopulation before.8 We consciously took steps to mini-mize any selection bias due to language barriers by usingthe three translations of WHOQOL-BREF.

According to the Malaysian National Renal Registrydata, median Spitzer’s QOL score (maximum score of 10)

of 9 and 10 were reported for 2011 in the Malaysian HDand PD patients respectively.1 It is unrealistic to reportperfect or near-perfect score in either modality. In allgeneric health-related QOL subscales, dialysis patientsfrom various continents reported much lower scores thantheir respective population norm values.15,16 Nevertheless,despite of lower overall QOL score, our cohort of PDpatients did significantly better than HD. By using theWHOQOL-BREF, Ginieri-Coccosis et al.17 also noted thatHD patients, compared with PD patients, may experiencereduced QOL after some years on therapy. Based on theSpitzer’s QOL index, Liu et al.18 found HD group is asso-ciated with reduced QOL compared with PD. In contrast,

Table 5 Standardized marginal mean of QOL and least significant difference (LSD) multiple test comparison of allsignificant factors that are associated with psychological domain

Variables Marginal means (95% CI) LSD multiple comparison test (P value)

Gender1. Male 57.7 (50.0, 65.3) Male > female (0.043)2. Female 55.3 (47.4, 63.1)

Religion1. Muslim 61.7 (54.0, 69.3) Muslim > Buddhist (<0.001)

Muslim > Hindu (0.004) Muslim > Christian (<0.001)Hindu > Buddhist (0.011)

2. Buddhist 52.7 (45.2, 60.2)3. Hindu 57.0 (49.0, 65.0)4. Christian 53.3 (44.8, 61.8)5. Others 57.7 (47.1, 68.4)

Education level1. Nil 54.2 (46.0, 62.3) Tertiary > nil (0.014)

Tertiary > primary (0.026)Tertiary > secondary (0.002)

2. Primary 56.2 (48.3, 64.2)3. Secondary 55.5 (47.8, 63.3)4. Tertiary 59.9 (51.9, 67.9)

Occupation1. Professional 57.6 (48.9, 66.3) Gen. worker > homemaker (0.025)

Gen. worker > retired (0.014)Homemaker > unemployed (<0.001)

2. Semi-professional 55.8 (47.4, 64.3)3. General worker 60.0 (51.8, 68.1)4. Laborer 58.2 (47.5, 69.0)5. Homemaker 55.3 (47.4, 63.2)6. Retired 55.3 (47.5, 63.1)7. Unemployed 53.1 (45.4, 60.7)

Hemoglobina 0.069 (0.005)DM

1. Yes 55.5 (47.8, 63.2) No > yes (0.049)2. No 57.5 (49.7, 65.3)

CHD1. Yes 55.1 (47.2, 63.1) No > yes (0.041)2. No 57.8 (50.2, 65.4)

CVA1. Yes 52.8 (43.9, 61.7) No > yes (0.008)2. No 60.1 (52.8, 67.4)

CI = confidence Interval; CHD = coronary heart disease; CVA = cerebral vascular accident; DM = diabetes mellitus; > = better QOL score than.aPresented with correlation coefficient (R-squared).



Boateng’s systematic review did not identify any signifi-cant difference of QOL between HD and PD.19 PD isknown to offer increased autonomy and control, flexibilityin lifestyle and the dietary regime, as well as fewer socialrestrictions.20 PD patients reported better QOL ratings inspecific areas like “perceived ability to travel,” “financialconcerns,” “restriction in eating and drinking,” and “dialy-sis access problems.”21 Furthermore, PD patientsexpressed more positive ratings in several disease QOLdomains, e.g., less kidney disease burden, and being moreencouraged and satisfied with care.22 HD patients indi-cated lower WHOQOL-BREF scores in the domains ofenvironment and social relationships compared with PD.23

Undoubtedly, HD and PD populations were distinct fromeach other, showing differences in terms of demography(higher proportion of males, Chinese, Buddhist, tertiaryeducation, unemployment in HD), and comorbid status

(higher proportion of DM, hypertension, visual impair-ment in PD; Table 1).

Religion and CVA are the only two independent, sig-nificant factors associated with QOL scores in all domains.Patel et al.24 examined the relationship between religiousbeliefs, QOL, and psychosocial measures in patients withESRD on dialysis. They found that religious beliefs wererelated to perception of depression, illness effects, socialsupport, and QOL independent of medical aspects ofillness. Religion may be associated with increased patientsatisfaction with life and increased levels of socialsupport.25 Spiritual and religious beliefs have been shownto have a significant impact on QOL and overall mortal-ity.24,26,27 This relationship has been well documented andstudied in the general population and in certain groupssuch as cancer survivors. In our cohort, it remains unclearhow a particular religion could impact on all four domains

Table 6 Standardized marginal mean of QOL and least significant difference (LSD) multiple test comparison of allsignificant factors that are associated with social domain


Religion1. Muslim 48.9 (43.4,54.3) Muslim > Buddhist (0.006)

Muslim > Hindu (<0.001)2. Buddhist 45.0 (39.5,50.5)3. Hindu 41.7 (35.5,47.9)4. Christian 44.7 (37.6,51.8)5. Others 46.1 (36.4, 55.8)

Marital status1. Single 42.5 (36.6,48.5) Married > single (<0.001)2. Married 49.3 (44.2,54.4)3. Widow/widower 46.7 (40.4,53.1)4. Divorced 42.5 (31.8,53.3)

Occupation1. Professional 47.0 (40.0, 54.1) Gen. worker > semi-professional (0.025)

Gen. worker > unemployed (0.001)Homemaker > unemployed (0.030)

2. Semi-professional 42.3 (35.0,49.6)3. General worker 49.1 (42.9,55.3)4. Laborer 45.4 (35.0,55.8)5. Homemaker 45.8 (39.7,51.8)6. Retired 45.0 (39.3,50.7)7. Unemployed 42.3 (36.8,47.8)

CHD1. Yes 43.3 (37.3,49.3) No > yes (0.010)2. No 47.2 (41.7,52.7)

CVA1. Yes 41.1 (33.4,48.7) No > yes (0.011)2. No 49.5 (44.5,54.4)

Leg amputation1. Yes 42.2 (34.8,49.6) No > yes (0.048)2. No 48.3 (43.3,53.4)

CI = confidence Interval; CHD = coronary heart disease; CVA = cerebral vascular accident; > = better QOL score than.

Liu et al.


of WHOQOL-BREF. This is further confounded by the factthat all Malays are Muslims. It is difficult to tease outwhether the effect on QOL is due to race or religion. Onepostulate is that the State Religious Department provides

good coverage of financial assistance including dialysiscost and living expenses to Muslims who are poor andneedy. There are no equivalent financial bodies to cater forthe other religious groups. The areas of spirituality and

Table 7 Standardized marginal mean of QOL and least significant difference (LSD) multiple test comparison of allsignificant factors that are associated with environment domain


Religion1. Muslim 62.2 (58.8,65.6) Muslim > Buddhist (<0.001)

Muslim > Hindu (0.005)Muslim > Christian (<0.001) Muslim > others (0.011)Hindu > Buddhist (<0.001)Hindu > Christian (0.012)

2. Buddhist 52.2 (48.7,55.6)3. Hindu 58.3 (54.3,62.3)4. Christian 52.5 (47.6,57.5)5. Others 53.8 (46.7,60.9)

Education level1. Nil 53.2 (48.7,57.6) Tertiary > nil (0.004)

Tertiary > primary (0.004)Tertiary > secondary (0.002)

2. Primary 55.0 (51.1,58.9)3. Secondary 55.5 (51.8,59.1)4. Tertiary 59.6 (55.6,63.5)

Hemoglobina 0.069 (0.005)CVA

1. Yes 53.3 (48.0,58.5) No > yes (0.033)2. No 58.3 (55.5,61.2)

CI = confidence Interval; CVA = cerebral vascular accident; > = better QOL score than.aPresented with correlation coefficient (R-squared).

Table 8 Standardized marginal mean of QOL and least significant difference (LSD) multiple test comparison of allsignificant factors that are associated with combined overall quality of life and general health

Variables (P value) Marginal means (95% CI) LSD multiple test comparison (P value)

BMI status1. Normal 64.0 (56.9,71.1) Normal > underweight (0.045)

Overweight > underweight (0.017)Overweight > obese (0.024)

2. Underweight 60.7 (53.0,68.4)3. Overweight 64.9 (57.6,72.1)4. Obese 61.4 (54.0,68.9)

Religion1. Muslim 68.3 (61.2,75.5) Muslim > Buddhist (<0.001)

Muslim > Christian (<0.001)Muslim > others (0.006)Hindu > Buddhist (<0.001)Hindu > Christian (0.007)Hindu > Others (0.022)

2. Buddhist 59.3 (52.3,66.3)3. Hindu 67.1 (59.6,74.6)4. Christian 60.6 (52.6,68.5)5. Others 58.4 (48.8,68.1)

Mode of dialysisHD 61.5 (54.5,68.6) PD > HD (0.031)PD 64.0 (56.6,71.4)

Hemoglobina 0.060 (0.004)CVA

1. Yes 59.0 (50.7,67.3) No > yes (0.004)2. No 66.5 (59.7,73.2)

BMI = body mass index; CI = confidence Interval; CVA = cerebral vascular accident; HD = hemodialysis; PD = peritoneal dialysis; > = betterQOL score than.aPresented with correlation coefficient (R-squared).



spiritualism have traditionally been omitted from conven-tional HRQOL instruments. Positive impact of spiritualityon QOL can be mediated through strong social support ofloved ones who care deeply for the patients’ holistic well-being. By caring, it allows the individual the opportunityto express concerns or fears in a neutral environment. Asa result, these positive reinforcements can improve one’shealth status or perceptions.27 We must recognize thatparameters that are related to spirituality and religiosityare meaningfully associated with patient’s perception ofQOL.28 Religiosity may play a role in acceptance of theillness based on human psychology but the issue of spiri-tuality or religious affiliation could be a stronger determi-nant which was not measured. Specific spiritualityinstruments can potentially tease out the differences ifpresent.

CVA was associated with poorer QOL in all fourdomains and overall QOL. Consequences of stroke such aspoor mobility, dependence on others for activities of dailyliving, unemployment, preoccupation with treatment, andlife expenses all may lead to losing psychological, physical,environmental29, and functional advantages.30

Research has documented the traditional factors thatpredicted good QOL. These include age less than 65years,31 female gender,32 higher socioeconomic31, and edu-cational status.26 We found females had poorer psycho-logical scores than males, a finding similar to the cohort of75 Indian HD patients, where female scored lower inpsychological and environmental domains of WHOQOL-BREF.33 Similar findings were found in Mingardi’s34 andSeica’s survey31 using Short Form 36. Educational level inour cohort was positively associated with psychologicaland environmental domains scores. Such findings weresimilarly reflected in Sathvik’s cohort of HD patients(n = 75) using the same QOL instrument.33

We found diabetic and its concomitant leg amputationswere independently associated with lower score in thephysical domain. A case control of an Iranian cohortshowed diabetic patients on HD having poorer physicaldomain score as compared with nondiabetics (P = 0.04).35

Advanced age together with degenerative comorbiditieshas been considered a significant determinant of poorQOL36 but that did not hold true in our cohort. Kimmelet al.37 showed that using the satisfaction with life scale,older HD patients are more satisfied with life even in theface of deteriorating physical function. Older people maycompensate better for deteriorating function by psycho-logical adjustment.38

The strength of this study is that the sample size hadadequate number of both HD and PD patients from thevarious sectors of dialysis providers. All potential subjects

from the participating centers were systematically derivedfrom the NRR database. The use of three translations ofWHOQOL-BREF facilitated participation of eligible sub-jects of all ethnic origin. No subjects were excluded fromthis study because of language barrier. However, there arelimitations of this study. We enlisted only big urban dialy-sis centers with experienced staff in clinical research toensure smooth running of the study protocol. Subjectsin rural centers are known to have a differentsocioeconomic-cultural profile. Dialysis patients maysuffer from the effects of uremia with compromised cog-nitive functions and memory deficits. The subjects’responses were all self-rated and subjected to recall bias.The WHOQOL-BREF version does not explore issuesrelated to spirituality or religiosity compared with the fullversion of WHOQOL.

In summary, we reported mean (standard deviation)transformed QOL scores were 56.2 (15.8), 59.8 (16.8),58.2 (18.5), 59.5 (14.6), 61.0 (18.5) for (1) physical, (2)psychological, (3) social relations, (4) environmentdomains, and (5) combined overall QOL and generalhealth, respectively. Independent factors that were associ-ated significantly with QOL score in different domainsinclude gender, body mass index, religion, education,marital status, occupation, income, mode of dialysis,hemoglobin, DM, CHD, CVA, and leg amputation. Ofnote, religion and CVA both were significantly associatedwith all domains and combined overall QOL and generalhealth.

ACKNOWLEDGMENTS

We would like to thank the Director General of Health,Malaysia for permission to publish this paper. The authorswish to thank Madam Lee Day Guat, manager of theMalaysian National Renal Registry, and her team and Mr.Adil Fazrul, clinical research officer from the InternationalIslamic University of Malaysia for their valuable assistancein this study.

The authors appreciate the voluntary participation ofthe nephrologists and paramedical staff from the studysites: (1) Kuala Lumpur General Hospital, Kuala Lumpur;(2) Sultanah Aminah Hospital, Johor Bahru; (3) TengkuAmpuan Rahimah Hospital, Klang; (4) Tuanku Jaafar Hos-pital, Seremban, (5) Charis-National Kidney FoundationDialysis Centre, Cheras; (6) Yayasan Kebajikan Sow SengLam, Puchong; (7) Ampang Puteri Hospital, KualaLumpur; (8) Seremban Specialist Hospital, Seremban;(9) University Malaya Medical Centre, Petaling Jaya;(10) University Malaya Specialist Centre, Petaling Jaya;(11) Haemodialysis Unit of University Kebangsaan

Liu et al.


Malaysia, Cheras; (12) Haemodialysis Unit of UniversityKebangsaan Malaysia, Bangi; (13) Premier Medical Centreof University Kebangsaan Malaysia, Cheras.

Manuscript received July 2013; revised September 2013.

REFERENCES1 Lim YN, Ong LM, Goh BL, eds. Nineteenth Report of the

Malaysian Dialysis and Transplant Registry 2011, KualaLumpur 2012.

2 Ifudu O, Paul HR, Homel P, Friedman EA. Predictivevalue of functional status for mortality in patients onmaintenance hemodialysis. Am J Nephrol. 1998; 18:109–116.

3 DeOreo PB. Hemodialysis patient-assessed functionalhealth status predicts continued survival, hospitalization,and dialysis-attendance compliance. Am J Kidney Dis.1997; 30:204–212.

4 Lim TO, Morad Z. Reliability, validity and discriminatoryability of Spitzer’s QL-index in dialysis patients. Med JMalaysia. 1998; 53:392–400.

5 World Health Organization. Measuring quality of life.1997. Available from: http://www.who.int/mental_health/media/68.pdf (accessed date: April 14, 2013).

6 Szabo S. The World Health Organization Quality of Life(WHOQOL) assessment instrument. In: Spiker B, ed.Quality of Life and Pharmacoeconomics in Clinical Trials.Philadelphia, PA: Lippincott-Raven. 1996; 353–362.

7 Development of the World Health OrganizationWHOQOL-BREF quality of life assessment. TheWHOQOL Group. Psychol Med. 1998; 28:551–558.

8 Hasanah CI, Naing L, Rahman AR. World Health Orga-nization Quality of Life Assessment: Brief version inBahasa Malaysia. Med J Malaysia. 2003; 58:79–88.

9 Hasanah Cl, Razali MS. The pilot study of WHOQOL-100 (Malay version). The Malay J Med Sci. 1999; 6:21–25.

10 The WHOQOL Group. Development of the WorldHealth Organization WHOQOL-BREF Quality of LifeAssessment. Psychol Med. 1998; 28:551–58.

11 World Health Organisation. June 1997. Available from:http://depts.washington.edu/seaqol/docs/WHOQOL-BREF%20with%20scoring%20instructions.pdf(accessed date: October 28, 2013).

12 World Health Organisation. 2004. Available from:http://www.who.int/substance_abuse/research_tools/en/chinese_whoqol.pdf (accessed date: October 28, 2013).

13 World Health Organization. WHOQOL-BREF: Introduc-tion, administration, scoring and generic version ofassessment. 1996. Available from: http://www.who.int/mental_health/media/en/76.pdf (accessed date: April 10,2013).

14 Cagney KA, Wu AW, Fink NE, et al. Formal literaturereview of quality-of-life instruments used in end-stagerenal disease. Am J Kidney Dis. 2000; 36:327–336.

15 Fukuhara S, Lopes AA, Bragg-Gresham JL, et al. Health-related quality of life among dialysis patients on threecontinents: The Dialysis Outcomes and Practice PatternsStudy. Kidney Int. 2003; 64:1903–1910.

16 Valderrabano F, Jofre R, Lopez-Gomez JM. Quality of lifein end-stage renal disease patients. Am J Kidney Dis. 2001;38:443–464.

17 Ginieri-Coccossis M, Theofilou P, Synodinou C, TomarasV, Soldatos C. Quality of life, mental health and healthbeliefs in haemodialysis and peritoneal dialysis patients:Investigating differences in early and later years ofcurrent treatment. BMC Nephrol. 2008; 9:14.

18 Liu WJ, Chew TF, Chiu AS, Zaki M. Quality of life ofdialysis patients in Malaysia. Med J Malaysia. 2006;61:540–546.

19 Boateng EA, East L. The impact of dialysis modality onquality of life: A systematic review. J Ren Care.. 2011;37:190–200.

20 Cameron JI, Whiteside C, Katz J, Devins GM. Differencesin quality of life across renal replacement therapies:A meta-analytic comparison. Am J Kidney Dis. 2000;35:629–637.

21 Wu AW, Fink NE, Marsh-Manzi JV, et al. Changes inquality of life during hemodialysis and peritoneal dialysistreatment: Generic and disease specific measures. J AmSoc Nephrol. 2004; 15:743–753.

22 Kutner NG, Zhang R, Barnhart H, Collins AJ. Healthstatus and quality of life reported by incident patientsafter 1 year on haemodialysis or peritoneal dialysis.Nephrol Dial Transplant. 2005; 20:2159–2167.

23 Theofilou P. Quality of life in patients undergoing hemo-dialysis or peritoneal dialysis treatment. J Clin Med Res.2011; 3:132–138.

24 Patel SS, Shah VS, Peterson RA, Kimmel PL. Psychosocialvariables, quality of life, and religious beliefs in ESRDpatients treated with hemodialysis. Am J Kidney Dis.2002; 40:1013–1022.

25 Koenig HG. MSJAMA: Religion, spirituality, and medi-cine: Application to clinical practice. JAMA. 2000;284:1708.

26 Mapes DL, Bragg-Gresham JL, Bommer J, et al. Health-related quality of life in the Dialysis Outcomes and Prac-tice Patterns Study (DOPPS). Am J Kidney Dis. 2004;44(Suppl 2):54–60.

27 Berman E, Merz JF, Rudnick M, et al. Religiosity in ahemodialysis population and its relationship to satisfac-tion with medical care, satisfaction with life, and adher-ence. Am J Kidney Dis. 2004; 44:488–497.

28 Kimmel PL, Emont SL, Newmann JM, Danko H, MossAH. ESRD patient quality of life: Symptoms, spiritualbeliefs, psychosocial factors, and ethnicity. Am J KidneyDis. 2003; 42:713–721.



29 Haghgoo HA, Pazuki ES, Hosseini AS, Rassafiani M.Depression, activities of daily living and quality of life inpatients with stroke. J Neurol Sci. 2013; 328:87–91.

30 Patel MD, McKevitt C, Lawrence E, Rudd AG, Wolfe CD.Clinical determinants of long-term quality of life afterstroke. Age Ageing. 2007; 36:316–322.

31 Seica A, Segall L, Verzan C, et al. Factors affecting thequality of life of haemodialysis patients from Romania:A multicentric study. Nephrol Dial Transplant. 2009;24:626–629.

32 Lew SQ, Patel SS. Psychosocial and quality of life issuesin women with end-stage renal disease. Adv ChronicKidney Dis. 2007; 14:358–363.

33 Sathvik BS, Parthasarathi G, Narahari MG, Gurudev KC.An assessment of the quality of life in hemodialysispatients using the WHOQOL-BREF questionnaire. IndianJ Nephrol. 2008; 18:141–149.

34 Mingardi G, Cornalba L, Cortinovis E, Ruggiata R,Mosconi P, Apolone G. Health-related quality of life indialysis patients. A report from an Italian study using theSF-36 Health Survey. DIA-QOL Group. Nephrol DialTransplant. 1999; 14:1503–1510.

35 Anees M, Hameed F, Mumtaz A, Ibrahim M, Saeed KhanMN. Dialysis-related factors affecting quality of life inpatients on hemodialysis. Iran J Kidney Dis.. 2011; 5:9–14.

36 Borglin G, Jakobsson U, Edberg AK, Hallberg IR. Olderpeople in Sweden with various degrees of present qualityof life: Their health, social support, everyday activitiesand sense of coherence. Health Soc Care Community.2006; 14:136–146.

37 Kimmel PL, Peterson RA, Weihs KL, et al. Aspects ofquality of life in hemodialysis patients. J Am Soc Nephrol.1995; 6:1418–1426.

38 Lamping DL, Constantinovici N, Roderick P, et al. Clini-cal outcomes, quality of life, and costs in the NorthThames Dialysis Study of elderly people on dialysis: Aprospective cohort study. Lancet. 2000; 356:1543–1550.

SUPPORTING INFORMATION

Additional Supporting Information may be found in theonline version of this article at the publisher’s web-site:

Word document on WHOQOL-BREF questionnaires inMalay, English, Chinese translations respectively.

Supplementary material is linked to the online versionof the paper at: http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1542-4758/issues

Liu et al.


quality of life in dialysis: a malaysian perspective

Documents