According to the World Malaria Report 2018, there were 219 million cases of malaria globally in 2017 (uncertainty range 203-262 million) and 435,000 malaria deaths [1]. In Indonesia itself, there were 261,617 cases of malaria in 2017, which killed at least 100 people. West Papua, which is included in the 3 major provinces with the highest malaria incidence, reported 9,110 malaria cases in 2016 [2].
The concept of QoL is very useful for the process of improving health, holistic therapy and care, and for the rehabilitation process [3]. Researches on QoL must take into account the patients’ somatic conditions, mental status, social relations and physical fitness. Health condition is one of the basic factors affecting QoL [4]. Malaria affects work productivity, with negative consequences for economic growth, which limits living conditions, supports the spread of disease in the population, which in turn will suffer a vicious circle of disease and poverty [5].
HRQOL could be measured using generic or specific instrument based on the purpose of measurement. The generic instrument is classified into health profiles and preference-based measures approach. Meanwhile, specific instrument measures the health profiles based on disease-specific, population-specific, function-specific, and condition- or problem-specific [6]. Example of specific instruments are Diabetes Quality of Life Clinical Trial Questionnaire-Revised (DQLCTQ-R) and Diabetes Specific Quality of Life Scale (DSQOLS) for measuring HRQOL of diabetes mellitus patients [7], the Dartmouth Cooperative Functional Assessment Charts (COOP) for measuring HRQOL in paediatrics [8], and Functional Outcomes of Sleep Questionnaire (FOSQ) for measuring HRQOL in obstructive sleep apnea [9]. Among the generic instruments to measure health profiles, Short Form (SF-36) is the most commonly used. The Quality of Well-Being (QWB) Scale, the Health Utilities Index (HUI), and the EuroQol Instrument (EQ-5D) are preference-based measures designed to summarise HRQOL in a single number ranging from 0 to 1, namely health utility (utility) [10]. Utility is necessary in economic evaluation/pharmacoeconomic studies to calculate quality of adjusted life years (QALY), which is an outcome used in cost utility analysis, one of method in pharmacoeconomic study. The concept of Quality Adjusted Life Years (QALYs) is a multiplication of HRQOL in term of utility and duration of life [11].
The aim of this study was to measure HRQOL on malaria patientsusing EQ-5D-5L instrument and examine the relationship between socio-demographic and HRQOL.
Materials and Methods
Study Design and Participants
This was a cross-sectional study, conducted from October to December 2018 (3 months) involving malaria patients who visited primary healthcare centres in Manokwari District, West Papua Province, Indonesia.
The inclusion criteria were a positive malaria diagnoses, age 18 years or above, and an ability to communicate well. The exclusion criteria were those who had serious diseases that might interfere QoL such as diabetes mellitus, chronic kidney disease, and cardiovascular disease; as well as pregnant mother. Sample size was determined using the following formula [6]:
![](/articles/images/13010//jcdr-13-IC01-e001.jpg)
n=sample size required (number of responders to survey)
σ=the standard deviation of utility mean from previous study
ω=precision or margin of error
1,96=the standard normal variate at 5% type 1 error (p<0.05)
The previous study of QoL measurement of malaria patients with 11 and 20 responders resulted Standard of Error (SE) of utility 0.003-0.061, hence the standard deviation were 0.13-0.26 [12]. Using the 0.05 (5%) precision, the sample size was calculated as 26-103 responders. In this study we recruited 110 malaria patients as study sample.
Study Site
Data were collected from four primary healthcare centresin three different areas of Manokwari Districs, which were Maripi from the South, Sanggeng and Amban from the West and Pasir Putih from the East. Study sites were selected based on the considerations of high prevalences of malaria in those healthcare centres and the accessibility of locations.
Research Instrument and Data Collection
The EQ-5D-5L instrument in Bahasa Indonesia language version, developed by EuroQoL group was applied in this study [13]. The instrument consists of two parts. The first, namely EQ-5D-5L descriptive system, consists of five dimensions: “mobility”, “self-care”, “usual Activity”, “pain/discomfort” and “anxiety/depression”. Each domain has five levels of possible response. For instance, mobility asking about walking, the response is either no problems (level 1), or slight problem (level 2), ormoderate problem (level 3), or severe problems (level 4), or unable to walk/extreme problems (level 5). Responses for the 5 domains is called the health state. For example, the health state “12345” indicates that there is no problem (level 1) in the dimension ofmobility, having slight problems in the dimension of usual activity, having moderate problems (level 3) in the domain of self-care, having severe problems (level 4) in the dimension of pain/discomfort, and having extreme problems (level 5) in the dimension ofanxiety/depression. The second part is the European Quality-Visual Analog Scale (EQ-VAS), consisting of the value of 0-100 units of the ‘thermometertype’ scale, where the value of 0 means died and 100 means perfect health [14].
Utility or EQ-5D index score was converted from EQ-5D-5L descriptive system using value set of Indonesia which represents Indonesian population [15].
Meanwhile VAS score was calculated using the following formula: the number respondent select on the ‘thermometertype’ scale, divided by 100.
An additional questionnaire consisted of patient sociodemographics, namely, age, gender, marital status, Education attainment, monthly income, occupation, type of malaria, family history, health problems, and health insurance. The patients were interviewed face-to-face by researchers using EQ-5D-5L and prepared questionnaire after signing an informed consent.
Statistical Analysis
All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) (version 21). Descriptive analysis included mean, standard deviations and frequencies, which were used to analyse socio-demographics of patients, EQ-5D-5L descriptive system, EQ-5D-5L index scores (utility), and EQ-VAS. Normality test of the data was tested using Kolmogorov-Smirnov test. Independent t-test was used to compare differences between mean of utility score among malaria patients. Chi-square test was performed to analyse cross tabulation between patient socio-demographics and EQ-5D descriptive system. Correlation between socio-demographics and utility score used spearman’s ranks correlation. Influence of socio-demographic on HRQOL were explored using multivariate regression analyses.
Ethical Considerations
This study obtained the ethical approval from Medical and Health Research Ethics Committee (MHREC) Faculty of Medicine Gadjah Mada University-Dr Sardjito General Hospital with the reference number of Ref: KE/FK/0472/EC/2018. The study began after receiving permissions from each study site of primary healthcare centres. Informed consents were collected from all study participants.
Results
Patients Socio-Demographics
Patient demographic characteristics are shown in [Table/Fig-1]. Out of 150 malaria patients recruited, 110 malaria patients met the inclusion criteria and were included in the study. The mean age of the patients was 33±12.7 years. The highest number of patients was in the age group 18-30 years (57.3%). Most of the respondent’s education level was below high school (63.6%). Approximately 49.1% of the respondents were employed. Most respondents were married (63.6%). Most of the respondents had a montly family income above Rp. 3,000,000 (56.4%). [Table/Fig-1].
Demographic characteristics malaria patients (2018, N=110).
Demographic characteristics | Number of patients (%) (N=110) |
---|
Gender | |
Male | 47 (42.7) |
Female | 63 (57.3) |
Age (years) | |
17-30 | 63 (57.3) |
31-50 | 36 (32.7) |
>50 | 11 (10) |
Mean±SD | 33.39±12.794 |
Educational attainment | |
<High school | 70 (63.6) |
>High school | 40 (36.4) |
Employment status | |
Employed | 54 (49.1) |
Unemployed | 56 (50.9) |
Income | |
<Rp. 3,000,000 | 48 (43.6) |
> Rp.3,000,000 | 62 (56.4) |
Marital status | |
Married | 70 (63.6) |
Single | 40 (36.4) |
Type of malaria | |
Severe | 40 (36.4) |
Mild | 70 (63.6) |
Family history of any disease | |
Yes | 31 (28.2) |
No | 79 (71.8) |
Other health problems | |
Yes | 45 (40.9) |
No | 65 (59.1) |
Health insurance | |
Yes | 103 (93.6) |
No | 7 (6.4) |
The types of malaria in the area of Manokwari Regency included mild malaria (63.6%) and severe malaria (36.4%). Severe malaria is defined as acute falciparum malaria (malaria tropica) with signs of severity and/or evidence of vital organ dysfunction. Mild malaria is of less severity and caused by Plasmodium ovale and Plasmodium vivax (Malaria tertiana), Plasmodium malariae (Malaria quartana) and Plasmodium knowlesi [16].
EQ-5D Descriptive System
Patients’ responses to five dimensions of EQ-5D-5L descriptive system were called health states. In this study we found 33 types of health states, with the highest health statebeing 22222, with 14 respondents (12.73%), followed by health state 33333, with 12 respondents (10.9%). The utility value for health state 22222 was 0.525 and for health state 33333 was 0.307 [Table/Fig-2].
Health state of HRQOL malaria patients measured with EQ-5D-5L.
Health state | Health utility | Number of patients | Percentage (%) |
---|
11121 | 0.914 | 4 | 3.6 |
11122 | 0.835 | 1 | 0.9 |
11221 | 0.824 | 5 | 4.5 |
12221 | 0.723 | 4 | 3.6 |
21221 | 0.705 | 9 | 8.2 |
21222 | 0.626 | 3 | 2.7 |
21223 | 0.571 | 1 | 0.9 |
21333 | 0.496 | 1 | 0.9 |
22211 | 0.690 | 1 | 0.9 |
22221 | 0.604 | 3 | 2.7 |
22222 | 0.525 | 14 | 12.7 |
22223 | 0.470 | 9 | 8.2 |
22232 | 0.516 | 6 | 5.5 |
22233 | 0.461 | 5 | 4.5 |
22323 | 0.404 | 1 | 0.9 |
22332 | 0.450 | 2 | 1.8 |
22333 | 0.395 | 3 | 2.7 |
23332 | 0.411 | 1 | 0.9 |
23333 | 0.356 | 1 | 0.9 |
32221 | 0.531 | 1 | 0.9 |
32322 | 0.386 | 2 | 1.8 |
32323 | 0.331 | 1 | 0.9 |
32331 | 0.456 | 1 | 0.9 |
32332 | 0.363 | 4 | 3.6 |
32333 | 0.322 | 2 | 1.8 |
32342 | 0.358 | 2 | 1.8 |
32433 | 0.138 | 1 | 0.9 |
33332 | .0338 | 3 | 2.7 |
33333 | 0.307 | 12 | 10.9 |
33342 | 0.235 | 2 | 1.8 |
33343 | 0.180 | 3 | 2.7 |
33443 | 0.035 | 1 | 0.9 |
44444 | -0.384 | 1 | 0.9 |
HRQOL: Health related quality of life; EQ-5D: European quality of life five-dimension. digits represent response level (1-5) for five dimensions (mobility, self-care, usual activities, pain/discomfort, anxiety/depression
[Table/Fig-3] presents responses for each domain of EQ-5D-5L descriptive system. Almost all of the respondents reported having problems in any of the domains. There were no responses for “extreme problems” for any of the domains. “Pain/discomfort” and “usual activities” domains were the most reported problems in malaria patients. Number of patients who reportedof having problems was 87.3% for “mobility”, 78.2% for “self-care”, 95.45% for “usual activities”, 99.1% for Pain/Discomfort, and 74.5% for “anxiety/depression”.
Response of some problems malaria patients with EQ-5D.
Dimension | N | % | Response of having problems (%) |
---|
Mobility |
No problems | 14 | 12.7 | 87.27 |
Slight | 61 | 55.5 |
Moderate | 34 | 30.9 |
Severe | 1 | 0.9 |
Unable/Extreme | 0 | 0 |
Self-care |
No problems | 24 | 21.8 | 78.18 |
Slight | 62 | 56.4 |
Moderate | 23 | 20.9 |
Severe | 1 | 0.9 |
Unable/Extreme | 0 | 0 |
Usual activities |
No problems | 5 | 4.5 | 95.45 |
Slight | 63 | 57.3 |
Moderate | 39 | 35.5 |
Severe | 3 | 2.7 |
Unable/Extreme | 0 | 0 |
Pain/Discomfort |
No problems | 1 | 0.9 | 99.09 |
Slight | 59 | 53.6 |
Moderate | 42 | 38.2 |
Severe | 8 | 7.3 |
Unable/Extreme | 0 | 0 |
Anxiety/Depression |
No problems | 28 | 25.5 | 74.54 |
Slight | 41 | 37.3 |
Moderate | 40 | 36.4 |
Severe | 1 | 0.9 |
Unable/Extreme | 0 | 0 |
From the results of the Chi-square test [Table/Fig-4], it can be seen that age significantly influenced patients’ reporting problems associated with QoL, particularly in the domains of “self-care” (p=0.039) and “anxiety/depression” (p=0.014). Frequency of “some problems” was significantly higher for mild malaria patients in the domains of “mobility” (p=0.002), “self care” (p=0.000), and “anxiety/depression” (p=0.000). It was found that patients with severe malaria did not have many problems in other dimensions. The frequency of respondents having problems was higher in the married group (p=0.002).
Patient response to problems in EQ-5D-5L (N=110).
| MO | SC | UA | PD | AD |
---|
% of having problems | p-value | % of having problems | p-value | % of having problems | p-value | % of having problems | p-value | % of having problems | p-value |
---|
Gender |
Female | 36.3 | 0.556 | 34.5 | 0.558 | 40.9 | 0.637 | 41.8 | 0.427 | 29.0 | 0.179 |
Male | 50.9 | 43.6 | 54.5 | 57.2 | 45.4 |
Age |
17-30 | 47.2 | 0.051 | 40 | 0.039* | 53.6 | 0.124 | 56.3 | 0.571 | 37.2 | 0.014* |
31-50 | 31.8 | 30 | 32.7 | 32.7 | 30 |
>50 | 8.18 | 8.1 | 9.09 | 10 | 7.2 |
Education attainment |
< High school | 55.4 | 0.957 | 49 | 0.727 | 60.9 | 0.601 | 62.7 | 0.636 | 48.1 | 0.710 |
> High school | 31.8 | 29 | 34.5 | 36.3 | 26.3 |
Employment status |
Employed | 43.6 | 0.617 | 41.8 | 0.081 | 48.1 | 0.364 | 49 | 0.509 | 37.2 | 0.744 |
Unemployed | 43.6 | 36.3 | 47.2 | 50 | 37.2 |
Income |
<Rp. 3,000,000 | 37.2 | 0.607 | 32.7 | 0.477 | 40.9 | 0.651 | 43.6 | 0.564 | 35.4 | 0.156 |
> Rp.3,000,000 | 50 | 45.4 | 54.5 | 55.4 | 39.0 |
Marital status |
Married | 57.2 | 0.256 | 52.7 | 0.116 | 61.8 | 0.351 | 63.6 | 0.364 | 53.6 | 0.002* |
Single | 30 | 25.4 | 33.6 | 35.4 | 20.9 |
Type of malaria |
Severe | 36.3 | 0.002* | 35.4 | 0.000* | 36.3 | 0.157 | 36.3 | 0.636 | 34.5 | 0.000* |
Mild | 56 | 42.7 | 59.0 | 62.7 | 40 |
Family history |
Yes | 25.4 | 0.753 | 24.5 | 0.156 | 27.2 | 0.565 | 28.1 | 0.718 | 24.5 | 0.058 |
No | 61.8 | 53.6 | 68.1 | 70.9 | 50 |
Health problems |
Yes | 35.4 | 0.874 | 32.7 | 0.701 | 38.1 | 0.398 | 40.9 | 0.591 | 33.6 | 0.124 |
No | 51.8 | 45.4 | 57.2 | 58.1 | 40.9 |
Health insurance |
Yes | 80.9 | 0.592 | 71.8 | 0.343 | 89.0 | 0.715 | 92.7 | 0.936 | 68.1 | 0.188 |
No | 6.3 | 6.3 | 6.3 | 6.36 | 6.3 |
EQ-5D-5L: European quality of life- 5 dimension- 5 level; MO: Mobility; SC: Self care; UA: Usual activity; PD: Pain/discomfort; AD: Anxiety/depression. p<0.05
p-value: chi-square test; No marks: not significant
EQ-5D Score and VAS Score
The EQ-5D score and VAS score based on socio-demographic groups are presented in [Table/Fig-5]. The mean EQ-5D-5L score patients was 0.49 ±0.20, while the mean VAS score was 0.509±0.117. The EQ-5D-5L score was slightly lesser than the VAS score. In this study, the mean EQ-5D score (0.349±0.126) and VAS score (0.434±0.088) in patients with severe malaria was significantly lower (p=0.000 and p=0.000 for EQ-5D score and EQ-VAS, respectively) compared to patients with mild malaria in which the mean EQ-5D scoreand EQ-VAS were 0.571±0.195 and 0.553±0.11, respectively.
HRQOL malaria patient with EQ-5D and VAS score. (N=110).
| EQ-5D score | VAS score |
---|
Mean | (SD) | p-value | Mean | (SD) | p-value |
---|
Total | 0.490 | 0.203 | | 0.509 | 0.117 | |
Gender | | | 0.734 | | | 0.430 |
Female | 0.482 | 0.207 | | 0.499 | 0.123 |
Male | 0.496 | 0.202 | | 0.517 | 0.113 |
Age | | | 0.161 | | | 0.171 |
17-30 | 0.516 | 0.227 | | 0.514 | 0.1170.108 |
31-50 | 0.437 | 0.145 | | 0.486 |
>50 | 0.516 | 0.205 | | 0.56 | 0.133 |
Education attainment | | | 0.689 | | | 0.539 |
< High school | 0.496 | 0.202 | | 0.514 | 0.122 |
> High school | 0.480 | 0.208 | | 0.50 | 0.109 |
Employment status | | | 0.220 | | | 0.418 |
Employment | 0.466 | 0.189 | | 0.50 | 0.113 |
Unemployment | 0.513 | 0.215 | | 0.518 | 0.121 |
Income | | | 0.979 | | | 0.681 |
<Rp. 3,000,000 | 0.489 | 0.224 | | 0.504 | 0.130 |
> Rp.3,000,000 | 0.490 | 0.188 | | 0.513 | 0.106 |
Marital status | | | 0.559 | | | 0.718 |
Married | 0.481 | 0.166 | | 0.512 | 0.107 |
Single | 0.505 | 0.257 | | 0.504 | 0.134 |
Type of malaria | | | 0.000* | | | 0.000* |
Severe | 0.349 | 0.126 | | 0.434 | 0.088 |
Mild | 0.571 | 0.195 | | 0.553 | 0.110 |
Family history of any disease | | | 0.607 | | | 0.665 |
Yes | 0.474 | 0.158 | | 0.517 | 0.117 |
No | 0.496 | 0.219 | | 0.506 | 0.117 |
Health problems | | | 0.674 | | | 0.158 |
Yes | 0.500 | 0.176 | | 0.528 | 0.104 |
No | 0.483 | 0.221 | | 0.496 | 0.124 |
Health insurance | | | 0.044* | | | 0.698 |
Yes | 0.500 | 0.191 | | 0.510 | 0.114 |
No | 0.341 | 0.321 | | 0.492 | 0.161 |
HRQOL: Health related quality of life; EQ-5D: European quality of life- 5 dimension; VAS: Visual analog scale; SD: Standard deviation. p< 0.05. Notes: independent t- test; No marks: not significant.
There was also significantly higher (p=0.044) mean EQ-5D score in patients having health insurance (0.500±0.191), compared to those not having health insurance (0.341±0.321). Nevertheless, there was no significant difference in the VAS score.
Results of multivariate regression analysis showed that two variables of the socio-demographics criteria, namely, type of malaria and health insurance, were found to be correlated with EQ-5D index score of malaria patients (p=0.05). These variables contributed to 36.9% of EQ-5D index score of malaria patients (R2=0.369) [Table/Fig-5,6].
Analysis with multivariate regression.
Variable | R | SE | Beta | t | p-value |
---|
Type malaria | -0.627 | 0.033 | -0.588 | -7.504 | 0.000* |
Health insurance | 0.108 | 0.065 | 0.308 | 3.932 | 0.000* |
Constant score | | 0.62 | | 5.527 | 0.000* |
R=0.607 R2=0.369 R2Adj=0.357 F=31.283 p-value=0.000* |
SE: Standard eror; t: t-test; p-value: Significant for variable; R: Multicorrelation coefficient; R2: Multiple regression correlation coefficient; R2Ad: Adjusted R value. *p<0.05
Discussion
To the best of our knowledge, this is the first study to measure HRQOL among malaria patients in Indonesia, particularly using EQ-5D-5L. We measured health status with EQ-5D descriptive system and converted to EQ-5D score using the value set for theIndonesian population.
In this study, all of respondents reported problems in any domain and, none of them reported full health state of 11111. Furthermore, the most reported domains of having problem were “pain/discomfort” (99%) and “usual activities” (95.4%). This indicated that malaria affected patient QoL due to “pain/discomfort” and since the patients were incapable of conducting “usual activities”. Other diseases might also affect patients ‘QoL through different domains. For example, the most reported problems in dengue fever were “self-care” (71.8%) and “usual activities” (64.6%) [17]. The most reported problems in cervical cancer were “pain/discomfort” (67.8%) followed by “anxiety/depression” (57.5%) [18]. The most reported problems in type 2 diabetes mellitus were “pain/discomfort” (76.33%) and “usual activities” (48%) [19]. The most reported problems in Indonesian general population were “pain/discomfort” (39.7%) and “anxiety/depression” (34.3%) [20].
McCarthy AE et al., reported that the HRQOL in terms of VAS score among malaria patients was 0.539(SE=0.039) [12], which was comparable to the VAS score among patients with simple malaria in Indonesia, that exhibited a VAS score of 0.509±0.117. The results of this study showed that utility (EQ-5D index score) among malaria patientsin Indonesia (0.49±0.20) was lowerthan other similar tropical diseases in other settings. For instance, the mean EQ-5D index score of Dengue Fever (DF) patients was 0.66 [17] and that of tuberculosis patients was 0.69 [21]. A study to measure QoL among malaria patients in Nigeria revealed that the majority of patients did not accept that they were suffering from malaria, which might have led to lower QoL [22].
The utility score among malaria patients was lower when compared to the utility of other diseases in Indonesia measured using the same instrument of EQ-5D, for instances the value of cervical cancer was 0.76±0.2 [17], while that of other, human papillomavirus (HPV) related cancers was 0.69±0.10 [23], and type 2 diabetes mellitus was 0.74±0.22 [19]. The utility of general population of Indonesia was 0.91 [20]. This indicates that the utility decrement of malaria patients from general population was about 0.42.
In this study, socio-demographic variables that significantly influenced QoL were age, marital status, and type of malaria severity. This finding was inline with a study conducted by Van Damme Ostapowicz K et al., [22] which indicates that age affects the social domain, especially in those over 50 years of age. Findings of the study conducted by McCarthy AE et al., showed that prevalence of severe malaria affects the time of hospitalisation when compared with mild malaria, which worsens patients’ QoL [12]. A study of conducted by Sakamaki H et al., in type 2 diabetes mellitus patients showed that the frequency of having problems was higher in patients with complications (more severe) than in those without complications (less severe) [24].
Limitation
This study is limited by the small sample size, which therefore, does not represent the whole of Indonesia, but only the Manokwari region. However, this study provides insight regarding HRQOL in malaria patients in a particular part of Indonesia.
Conclusion
This study focuses on measuring the QoL in malaria patients in Manokwari, Indonesia. Malaria was associated with poor HRQOL and particularly affected “Pain/discomfort” and “usual activities” domains. The type (severity) of malaria and health insurance were found to be correlated with QoL on malaria patients. These findings suggest that, measurement of QoL in malaria patients using EQ-5D-5L provides an overview of their own health status and it also helps health workers to provide care to patients and focus on some problem domains.
Fundings: This research was supported by the Ministry of Research, Technology and Higher Education Republic of Indonesia.