Investigating validity evidence of the Malay translation of the Copenhagen Burnout Inventory.

INTRODUCTION: The Copenhagen Burnout Inventory (CBI) is a recent burnout measure with a focus on fatigue and exhaustion. It has three factors: personal burnout, work-related burnout, and client-related burnout. This study aimed to translate the CBI into the Malay language and to validate the translated version among a group of medical students.
METHODS: The forward-backward translation was performed as per standard guidelines. The Malay version of CBI (CBI-M) was distributed to 32 medical students to assess face validity and later to 452 medical students to assess construct validity. The data analysis was performed by Microsoft Excel, SPSS and AMOS.
RESULTS: The face validity index of CBI-M was more than 0.8. The three factors of CBI-M achieved good levels of goodness-of-fit indices (Cmin/df = 2.99, RMSEA = 0.066, GFI = 0.906, CFI = 0.938, NFI = 0.910, TLI = 0.925). The composite reliability values of the three factors ranged from 0.84 to 0.87. The Cronbach's alpha values of the three factors ranged from 0.83 to 0.87.
CONCLUSIONS: This study supports the face and construct validity of the CBI-M with a high internal consistency.

Introduction

Since the conceptualization of burnout, three main burnout inventories have been developed to assess burnout,1, 2, 3 which include the Maslach Burnout Inventory (MBI), the Oldenburg Burnout Inventory (OLBI)5, 6 and the Copenhagen Burnout Inventory (CBI); the MBI remains the ‘gold standard’ to assess burnout.2, 3 The MBI was developed based on the original three-dimensional conceptualization of burnout, which incorporated depersonalization and reduced sense of personal accomplishment (i.e., inefficacy).2, 3, 4, 8, 9 The OLBI was developed based on exhaustion and disengagement (i.e., depersonalization) dimensions, and the exhaustion dimension was expanded to include cognitive and physical exhaustion, in addition to the MBI's presumed focus on emotional exhaustion.1, 2, 3, 6, 10 The CBI is the newest tool developed to assess burnout, and it claimed to assess the core features of burnout, i.e., fatigue and exhaustion, in relation to personal life (i.e., personal burnout), work (i.e., work-related burnout) and service to clients (i.e., client-related burnout).1, 3, 7 Apart from that, being a public domain instrument is an advantage of CBI (and OLBI) over MBI.5, 7, 11, 12 It is noteworthy that although, in practice, the conceptualization of burnout is used by a majority of researchers, not all of them mean the same thing when they refer to ‘burnout’.1, 3

CBI's personal burnout category refers to the degree of physical and psychological fatigue and exhaustion experienced by a person. The items of personal burnout are generic questions, and therefore all participants will be able to answer them. CBI's work-related burnout category refers to the degree of physical and psychological fatigue and exhaustion that is perceived by a person in relation to work. The items of work-related burnout are more specific and focus on a person's burnout symptoms that are related to work. CBI's client-related burnout category refers to the degree of physical and psychological fatigue and exhaustion that is perceived by a person in relation to work with clients. The ‘client’ is a general term covering people such as patients, students, teachers, children, etc., who receive service (i.e., service recipients) from people who provide the service (i.e., service providers). The items of client-related burnout specifically assess the connection between fatigue and people-centred work.

Since its inception, at least eight studies have provided substantial evidence to support the CBI's validity in terms of content, response process (i.e., clear and easy to understand), internal structure (i.e., construct and internal consistency) and relations to other variables such as mental health and vitality.7, 11, 13, 14, 15, 16, 17, 18 In addition, the CBI has been translated into several major languages (English, Mandarin, Cantonese, Japanese, Swedish, Finnish, French and Slovenian) and has been validated in Denmark, Australia, Taiwan, New Zealand, Portugal, Brazil, Spain, Hong Kong, Italy and Serbia. However, the CBI has not been translated or validated into the Malay language, a language of Austronesian origin that is widely spoken in Southeast Asia and beyond.

Our study aimed to produce a valid Malay translation of the CBI (CBI-M) to measure burnout among the Malaysian population. As previously mentioned, unlike the MBI, the CBI is freely available in the public domain11, 12; hence, from the cost-benefit perspective, it is suitable to be used by students, teachers and administrators for assessment, training and research purposes. This, in turn, will promote more burnout research in Malaysia. Building on this purpose, this study was designed to answer three research questions: 1) Are the items of the CBI-M able to be understood clearly and easily by Malaysian respondents? 2) Do the three factors of the CBI-M achieve a satisfactory level of construct validity? 3) Do the three factors of the CBI-M show a high level of internal consistency?

Materials and Methods

The forward-backward translation of CBI

The forward–backward translation technique was performed based on the recommended translation guidelines. FI, a psychiatrist (a content expert), and NNH, a professional linguistics teacher (a language expert), translated the original English version of the CBI into the Malay version (Forward Translation). A meeting was then held to reconcile and finalize the Malay version (CBI-M). RZ, another psychiatrist (a content expert), and SAMK, another professional linguistics teacher (a language expert), translated the Malay version of the questionnaire back into an English version (Backward Translation). This was followed by a meeting to reconcile the translated and original English versions of the CBI. Figure 1 illustrates the details of the translation process. The CBI-M is provided in the Appendix.

Figure 1

Translation and validation process of CBI.

The validation study procedure

The CBI-M was distributed to 452 medical students based on the recommended ratio of 10–20 samples per item for a validation study. Inclusion criteria were Malaysian medical students aged 18 years old and above. They were proficient in the Malay language and agreed to participate in the study. Exclusion criteria were non-Malaysian medical students, students who did not give their consent and students who were not proficient in the Malay language.

Eligible participants were provided with an information sheet that contained relevant details of the study, and informed consent was obtained. Following this, demographic details of participants were recorded.

Participants were approached individually via Facebook Messenger through their Facebook account. The data were collected through an online questionnaire developed using Google Forms. They received an informed consent form reassuring them about anonymity, confidentiality, and that published results were solely for scientific purposes. Due to the use of online links, all the attempted questionnaires were completed by the participants.

The face validity of the CBI-M was assessed by measuring its clarity and comprehensibility by 32 medical students from the same institution who were not involved with the construct validity study.

Subsequently, the construct validity of the CBI-M was tested on 452 medical students who were not involved in the face validity study. Out of 452, 115 were 1st-year medical students, 85 were 2nd-year medical students, 76 were 3rd-year medical students, 88 were 4th-year medical students and 88 were 5th-year medical students.

Ethical approval

Ethical approval was obtained from the Human Research Ethics Committee Universiti Sains Malaysia (HREC).

Data and statistical analysis

Data analysis was performed by Microsoft Excel, Statistical Package for the Social Sciences (SPPS) version 22 and Analysis of Moment Structures (AMOS) software.

Face validity

Two indices of face validity – clarity and comprehension – were assessed through 5-point Likert-scale responses collected from 32 medical students based on the CBI-M items using SPSS. For clarity, the participants were requested to respond on a scale of 1 (not clear at all) to 5 (very clear), and for comprehension on a scale of 1 (unable to understand at all) to 5 (easily understood). These responses were then categorized into 0 (not clear or unable to understand) and 1 (clear or able to understand) for calculation of face validity index. The universal face validity index was calculated by averaging the index values of clarity and comprehension. In this study, the value of the content validity index was adopted to interpret the value of the face validity index, in which 80% and above was considered as a satisfactory level of face validity.

Construct validity

The construct validity of the CBI-M was assessed through goodness of fit indices, convergent validity and discriminant validity. The latent constructs of the CBI-M were considered to fit if all the goodness-of-fit indices achieved the minimal requirements, as stated in Table 1. Convergent validity was checked with size of factor loading, average variance extracted (AVE) and composite reliability (CR). For each construct, item factor loading values should be reasonably high (0.5 or more) to signify convergent validity. AVE and CR were calculated manually using the formulae as recommended by Fornell and Larcker (1981) and Hair et al. (2009). Values of 0.5 or more for AVE and 0.6 or more for CR were considered as indicators to signify convergent validity.22, 23, 24 Discriminant validity was tested by comparing its AVE and shared variance (SV) values. SV is given as the square of correlation between two constructs. Constructs were considered to have achieved acceptable levels of discriminant validity when their AVE values were higher than SV values. A correlation of more than 0.85 between constructs was considered as an indicator to signify poor discriminant validity.

Table 1

Goodness-of-fit indices that were used to signify model fit.

Name of category	Name of index	Level of acceptance
Absolute fita	Root Mean Square of Error Approximation (RMSEA)	Less than 0.0830
	Goodness-of-Fit Index (GFI)	More than 0.931
Incremental fitb	Comparative Fit Index (CFI)	More than 0.932
	Tucker–Lewis Index (TLI)	More than 0.933
	Normed Fit Index (NFI)	More than 0.934
Parsimonious fitc	Chi-Squared/Degree of Freedom (Chisq/df)	Less than 535

Absolute fit: Measures overall goodness of fit for both the structural and measurement models collectively. This type of measure does not make any comparison to a specified null model (incremental fit measure) or adjust for the number of parameters in the estimated model (parsimonious fit measure).

Incremental fit: Measures goodness of fit that compares the current model to a specified “null” (independence) model to determine the degree of improvement over the null model.

Parsimonious fit: Measures goodness of fit representing the degree of model fit per estimated coefficient. This measure attempts to correct for any “overfitting” of the model and evaluates the parsimony of the model compared to the goodness of fit.

Internal consistency

Internal consistency would be determined by reliability analysis using SPSS, as reflected by Cronbach's alpha coefficient. Cronbach's alpha values between 0.7 and 0.9 were considered to show high internal consistency, and values between 0.6 and 0.7 were considered to show satisfactory internal consistency.

Results

Out of 452 respondents, 167 (36.9%) were males and 285 (63.1%) were females. Pertaining to ethnic groups, 195 (43.1%) were Malay, followed by 174 Chinese (38.5%), 66 Indian (14.6%) and 17 from other groups (3.8%).

Face validity of CBI-M

The face validity index of clarity and comprehension were 82.7% and 85.9%, respectively, and the universal face validity index was 83.2%, indicating a satisfactory level of face validity. The details of item-level indices are shown in Table 2. For the personal burnout (PB) factor (item: 1a, 1b, 1c, 1d, 1e, 1f), the item face validity index (universal) ranged between 78.1% and 92.2%. For the work-related burnout (WR) factor (item: 2a, 2b, 2c, 2d, 2e, 2f, 2g), the item face validity index (universal) ranged between 78.1% and 90.7%. For the client-related burnout (CR) factor (item: 3a, 3b, 3c, 3d, 3e, 3f), the item face validity index (universal) ranged between 78.2% and 95.4%. These results support a satisfactory level of face validity at the item level. Construct Validity of CBI-M showed that the one-factor model with 19 items was not a model fit, indicating that the CBI-M has multiple constructs. The original three-factor model with 19 items also failed to achieve acceptable values of the goodness-of-fit indices for the CBI-M, suggesting poor model fit.

Table 2

The face validity index of clarity and comprehension by CBI-M items.

No.	Question	Face validity index, % agreement
		Clarity	Comprehension	Universal
1(a).	Berapa kerap anda berasa letih?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	78.1	78.1	78.1
1(b).	Berapa kerap anda berasa letih secara fizikal?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	75.0	84.4	79.7
1(c).	Berapa kerap anda berasa letih secara emosi?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	78.1	84.4	81.3
1(d).	Berapa kerap anda berfikir: “Saya sudah tidak sanggup meneruskannya”?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	90.6	93.8	92.2
1(e).	Berapa kerap anda berasa lesu?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	84.4	81.3	82.9
1(f).	Berapa kerap anda berasa lemah dan senang mendapat penyakit?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	75.0	81.3	78.2
2(a).	Adakah kerja anda meletihkan emosi anda?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	78.1	81.3	79.7
2(b)	Adakah anda berasa lesu upaya (burnout) disebabkan pekerjaan anda?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	84.4	90.6	87.5
2(c).	Adakah anda berasa kecewa dengan pekerjaan anda?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	93.8	87.5	90.7
2(d).	Adakah anda berasa lesu pada akhir hari bekerja?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	75.0	87.5	81.3
2(e).	Adakah anda keletihan pada waktu pagi apabila memikirkan sehari lagi di tempat kerja?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	78.1	81.3	79.7
2(f).	Adakah anda berasa setiap waktu bekerja memenatkan anda?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	78.1	84.4	81.3
2(g).	Adakah anda mempunyai tenaga yang mencukupi untuk keluarga dan rakan rakan pada masa lapang?[0] Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	78.1	78.1	78.1
3(a).	Adakah anda mengalami kesukaran untuk berurusan dengan klien?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	96.9	93.8	95.4
3(b).	Adakah anda berasa berurusan dengan klien mengecewakan?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	90.6	87.5	79.7
3(c).	Adakah bekerja dengan klien menghabiskan tenaga anda?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	93.8	90.6	81.3
3(d).	Apabila bekerja dengan klien, adakah anda berasa lebih banyak memberi daripada menerima?[0) Pada tahap yang sangat tinggi 1) Pada tahap yang tinggi 2) Agak tinggi 3) Pada tahap yang rendah 4) Pada tahap yang sangat rendah]	71.9	84.4	92.2
3(e).	Adakah anda bosan bekerja dengan klien?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	84.4	87.5	82.9
3(f).	Adakah anda tertanya-tanya sejauh mana anda mampu meneruskan urusan dengan klien?[0) Sentiasa 1) Kerap kali 2) Kadang-kadang 3) Jarang-jarang 4) Tidak pernah/Hampir tidak pernah]	87.5	93.8	78.2
Average		82.7	85.9	83.2

Personal burnout (PB) = 1a, 1b, 1c, 1d, 1e, 1f; Work-related burnout (WR) = 2a, 2b, 2c, 2d, 2e, 2f, 2g; Client-related burnout (CR) = 3a, 3b, 3c, 3d, 3e, 3f.

Stepwise removal of items was performed based on modification indices, standardized residual covariance and standardized regression weighted to improve the model fit. The model fit was achieved after removal of 1 item that resulted in the three-factor model with 18 items, as shown in Table 3.

Table 3

The results of confirmatory factor analysis of the CBI-M.

Variable	X2 – statistic (df)	p-Value	Goodness-of-fit indices
			Cmin/df	RMSEA	GFI	CFI	NFI	TLI
One-factor modela	981.96 (152)	<0.001	6.46	0.110	0.760	0.799	0.771	0.774
Three-factor modela	518.80 (149)	<0.001	3.48	0.074	0.879	0.910	0.879	0.897
Three-factor modelb	383.20 (128)	<0.001	2.99	0.066	0.906	0.938	0.910	0.925

Based on the proposed construct; 19 items.

Based on the final model; 18 items – Item 2g was removed from the original construct.

All the goodness-of-fit indices were achieved to signify model fitness of the three-factor model. The final model of the CBI is illustrated in Figure 2, and details of standardized factor loading values are summarized in Table 4.

Figure 2

Standardized factor loading of CBI constructs based on the final model.

Table 4

The reliability analysis of the 18-item CBI-M based on the final model.

Item	Standardized factor loading	bDomain	aCronbach's alpha	cAVE	dCR
1(a). Berapa kerap anda berasa letih?	0.74	PB	0.85	0.48	0.85
1(b). Berapa kerap anda berasa letih secara fizikal?	0.70
1(c). Berapa kerap anda berasa letih secara emosi?	0.68
1(d). Berapa kerap anda berfikir: “Saya sudah tidak sanggup meneruskannya”?	0.65
1(e). Berapa kerap anda berasa lesu?	0.79
1(f). Berapa kerap anda berasa lemah dan senang mendapat penyakit?	0.60
2(a). Adakah kerja anda meletihkan emosi anda?	0.72	WR	0.87	0.52	0.87
2(b). Adakah anda berasa lesu upaya (burnout) disebabkan pekerjaan anda?	0.80
2(c). Adakah anda berasa kecewa dengan pekerjaan anda?	0.63
2(d). Adakah anda berasa lesu pada akhir hari bekerja?	0.68
2(e). Adakah anda keletihan pada waktu pagi apabila memikirkan sehari lagi di tempat kerja?	0.75
2(f). Adakah anda berasa setiap waktu bekerja memenatkan anda?	0.73
3(a). Adakah anda mengalami kesukaran untuk berurusan dengan klien?	0.70	CR	0.83	0.47	0.84
3(b). Adakah anda berasa berurusan dengan klien mengecewakan?	0.77
3(c). Adakah bekerja dengan klien menghabiskan tenaga anda?	0.78
3(d). Apabila bekerja dengan klien, adakah anda berasa lebih banyak memberi daripada menerima?	0.51
3(e). Adakah anda bosan bekerja dengan klien?	0.70
3(f). Adakah anda tertanya-tanya sejauh mana anda mampu meneruskan urusan dengan klien?	0.62

Item 2g was removed from the original construct.

Reliability analysis; Cronbach's alpha coefficient, overall Cronbach's alpha = 0.93.

Domains were predetermined based on a previous study.

AVE (average variance extracted) was calculated manually.

CR (composite reliability) was calculated manually.

The reliability analysis, as shown in Table 4, also confirmed the high level of internal consistency in the final model, with overall Cronbach's alpha being more than 0.7. The Cronbach's alpha of individual CBI-M factors ranged between 0.83 and 0.87, suggesting high internal consistency in these factors.

In addition, the composite reliability values of the CBI-M constructs ranged between 0.84 and 0.87, which indicated a high level of convergent validity (Table 4). Furthermore, all the standardized factor loading was more than 0.5, suggesting an adequate level of convergent validity. Table 5 shows that most of the AVE values of each factor are more than its SV values, indicating a good level of discriminant validity.

Table 5

AVE and SV of CBI based on the final model.

Factors	AVE	SV by factor
		PB	WR	CR
PB	0.48	1	0.51	0.18
WR	0.52	0.51	1	0.40
CR	0.47	0.18	0.40	1

Discussion

The translation process is thorough, as it followed the standard recommendation of forward and backward translation guidelines and involved both content experts and language experts. In addition, the reconciliation meetings held between researchers, content experts and language experts to finalize the translation product enhanced the rigour of translation process.

All items of the CBI-M were easily understood by participants, as evidenced by the face validity index values, which were more than 80% (Table 2), similar to a previously published study, although that study population was different (employees working in human resources services). We found that it was hard to find appropriate words for different expressions of burnout, and the terms ‘tired,’ ‘physically exhausted,’ ‘psychologically exhausted,’ and ‘worn-out’ all sounded the same in the Malay language. This effect was similarly reported in a previous study, but for the Chinese language. It suggests that cultural differences in expression of burnout are important and should be considered in future studies. Despite the challenges faced during the translation process, the CBI-M seems to have a satisfactory level of face validity, and it is considered to be an appropriate response process that evidence validity.

Similar to findings of previous studies,7, 11, 13, 14, 16, 17 the three factors of the CBI-M have shown good factor structures, as evidenced by their model fitness. All goodness-of-fit indices also supported its construct validity – convergent and discriminant validity – after removal of one item (Do you have enough energy for family and friends during leisure time?). To shed some light on why removal of one item supported validity of the CBI-M, two studies also found the same item had posed similar validity issues. The same item was interpreted and understood differently by men and women, suggesting the possibility of gender bias. In addition, the item was poorly loaded on the work-related factor and highly correlated with work frustration rather than work-related burnout; therefore, this warrants further investigation. Despite the unfavourable finding on this particular item, we opted to retain the item for future uses of the CBI-M in order to facilitate comparison of burnout scores across cultures or contexts, which is in line with the recommendation by Fong et al. (2014). Apart from that, the acceptable correlation between the three factors of the CBI-M (i.e., less than 0.85, as illustrated in Figure 2), except for personal burnout and work-related burnout, suggests that the CBI-M items were measuring different attributes of burnout. Similar to the finding of Fong et al. (2014), the high correlation (i.e., more than 0.85, as illustrated in Figure 2) between personal burnout and work-related burnout suggests a substantial overlap between these two domains. Even more, Milfont et al. (2008) found that the three factors of the original English CBI were strongly correlated with each other, hence suggesting the possibility of having a general burnout factor. Nevertheless, this present study and other studies7, 13, 14, 16, 17 showed that the three-factor model displays a significantly better fit than either the one-factor or two-factor model, supporting differentiation of the three domains of burnout.

The high internal consistency of the CBI-M is consistent with other reported studies.7, 11, 13, 14, 15, 16, 17, 18 Uniform findings on the internal consistency of the CBI in different countries, settings and languages indicate its internal structural stability. These findings further support the use and applicability of this public domain measure in broader contexts than is originally proposed by Kristensen et al. (2005).

Despite the challenges discussed above, given the adequate psychometric properties of the new CBI-M, we agreed with Yeh et al. and Campos et al. that the CBI is best for assessment of fatigue/exhaustion and does not confuse the experiences of burnout with other components, such as coping strategies or consequences of the syndrome.13, 14 Due to this reason, the CBI facilitates identification and clarification of causal relationships of burnout, provides better understanding of burnout and helps to plan interventions to minimize unwanted consequences of burnout at the personal or professional level.

There are limitations to this study. This study was confined to a group of medical students in a Malaysian medical school, and therefore future studies should involve other professions in order to verify the psychometric credentials of the CBI-M. Additional research should also investigate the association between the CBI-M and other, more well-established measures of burnout (i.e., the Maslach Burnout Inventory) to provide more evidence to support its validity. Last but not least, this study provided only validity evidence that is related to response process and internal structure; hence, future research is needed to investigate other sources of evidence to further support its validity. These include relationships of CBI-M scores with other relevant outcomes (i.e., mental health, quality of life, and psychological health) and their consequences on important outcomes (i.e., employability, physical illness, and medical errors).

Notwithstanding the above limitations, our study findings support the use of CBI-M – a free-of-charge burnout inventory – to assess burnout among the Malaysian population. It is also worth mentioning that, despite the many versions of the CBI in eight different languages and tested over 15 different occupational groups,7, 11, 13, 14, 15, 16, 17, 18 this is the first study that presents the Malay version of the instrument and applied it to a sample of medical students.

Conclusion

The CBI-M is a valid burnout tool with good psychometric properties tested in a group of medical students. Being a flexible and versatile tool, the CBI-M is recommended for future research to assess burnout in Malaysian settings, especially with its exclusive focus on fatigue and exhaustion.

Funding

This study was not funded by any grants.

Conflict of interest

The authors have no conflict of interest to declare.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institution (ethical code approval: USM/JEPeM/15020029).

Informed consent

Informed consent was obtained from all individual participants included in the study.

Authors' contribution

RAWC, YYC, MNC, NFM, MSW, YYL, and MSBY conceived and designed the study, provided research materials, collected and organized the data, analysed and interpreted data, wrote the initial draft and finalized the article, critically reviewed the final draft, and are responsible for the content and similarity of the manuscript.