https://orcid.org/0000-0002-6263-256X
Figures
Abstract
Concerns about the health and safety of working populations as well as preventive actions to reduce heat-related illnesses and fatalities have intensified as global warming and heatwaves continue to rise as a result of climate change. This study aimed to translate and culturally-adapted the translated Malay version of the Heat Strain Score Index (HSSI) questionnaire so that it can be utilized as a screening tool for heat stress among the Malay-speaking outdoor workers. The original English version of HSSI underwent forward-backward translation and was cross-culturally adapted into the Malay language by bilingual translators based on established guidelines. The content validation was reviewed by a six-member expert committee including the representative of outdoor workers. Face validation was carried out among 10 outdoor workers involved with various work tasks. Psychometric analysis was conducted based on a cross-sectional study among 188 workers who were eligible. Exploratory Factor Analysis (EFA) was used for construct validity while internal consistency reliability was performed using Cronbach’s alpha. The interclass correlation coefficient (ICC) was used to calculate the test-retest reliability. Both content and face validity were acceptable with the overall content validity index being 1.00, while the universal face validity index was 0.83. The factor analysis using varimax rotation extracted four factors which explained 56.32% of the cumulative percentage of variance and factor loading ranging from 0.415 to 0.804. The internal consistency reliability was acceptable with Cronbach’s alpha ranging from 0.705 to 0.758 for all the factors. The overall ICC value was 0.792 (95% CI; 0.764–0.801) which signifies good reliability. The findings from this study indicate that the Malay version of HSSI is a reliable and culturally-adapted instrument. Further validation is needed so that it can be used extensively assess the heat stress among susceptible Malay-speaking outdoor workers in Malaysia who are exposed to hot humid environments.
Citation: Lim MC, Lukman KA, Giloi N, Jeffree MS, Saupin SS, Sidek ZN, et al. (2023) Translation and cross-cultural adaptation of heat strain score index (HSSI) into the Malay language. PLoS ONE 18(2): e0281217. https://doi.org/10.1371/journal.pone.0281217
Editor: Nazarudin Safian, Universiti Kebangsaan Malaysia, Faculty of Medicine, MALAYSIA
Received: September 10, 2022; Accepted: January 18, 2023; Published: February 22, 2023
Copyright: © 2023 Lim et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
As global warming and heat waves continue to rise due to climate change, concerns regarding the health and safety of working populations, as well as preventive measures to minimize heat-related illnesses and fatalities, have increased. Outdoor workers are most susceptible to heat-related illnesses due to the challenging nature of their work, performed mainly outdoors under poor working conditions at temperatures over 35°C [1]. Heat stress not only increases the risk of heat-related illnesses but also intensifies the risk of accidental occupational injuries as a result of dizziness, exhaustion, lack of concentration, impaired decision-making or judgment, muscle weaknesses as well as clammy hands, and foggy safety goggles from the hot, humid working environment [2–4].
According to the National Institute of Occupational Safety and Health (NIOSH), heat stress is defined as the accumulation of heat the body experiences, which is measured as the total of the heat generated in the body (metabolic heat) in addition to environmental heat gained from the surroundings including work settings, but deducting the heat lost from the body to the environment [5]. A considerable radiant heat load can further aggravate heat stress from direct sunbeams, internal body heat, and an increased metabolic rate, all of which are caused by physical and intense activity [4].
Most workers and employers in Malaysia still lack awareness of the potential hazard of working in hot weather, especially as the ambient temperature continues to rise due to climate change [6]. According to an online survey conducted by Zander & Mathew (2019) among Malaysian adults, almost 99% of the 514 respondents claimed to experience heat stress symptoms at some point, which resulted in a 50% reduction in work capacity and a 29-days per year reduction in productivity [7]. Hence, employers must understand and prepare mitigating measures to face climate change that may affect their employees’ health, safety, and welfare in accordance with the Occupational Safety and Health Act (OSHA) 1994 [8].
A heat stress index incorporates a few elements, namely personal, physiological, and thermal environmental parameters, into a single value to measure workers’ heat stress exposure [9, 10]. Many heat indices have been recommended for various hot and humid environments, but no single heat index has gained universal recognition based on the various parameters involved [11, 12]. Heat indices, which are low-cost, non-invasive, simple to use, as well as based on an individual’s subjective experience, observation, and self-reported symptoms, have also been developed to assess heat stress responses [13, 14]. These indicators are thought to be essential in overcoming the limitations of invasive and precise field measurements.
One of them is the heat strain score index (HSSI), a questionnaire developed to determine the risk factors for heat strain [15, 16]. The individual’s observations, perceptions, and judgments about heat-related symptoms and behaviours are used to calculate this index [15, 16]. HSSI has been frequently applied over the years to analyze and compare heat strain responses in a variety of occupations, including bakers [17], municipal workers [18], open-pit miners [19], and steel workers [20].
Despite its widespread use, there is no translation performed on the HSSI questionnaire into the Malay language that has been published to the author’s knowledge. The translated Malay version HSSI questionnaire is important to assess the workers’ perception and observation of their work environment as well as their perceived symptoms at work which will also assist in the control and preventive measures at work. The main objective of this research is to translate, adapt and validate the translated Malay version of the HSSI questionnaire so that it can be utilized as a screening tool for heat stress and heat strain among the Malay-speaking high-risk working population in Malaysia before a further invasive assessment is conducted.
Materials and methods
HSSI questionnaire
Validated Heat Strain Score Index (HSSI) to assess responses based on respondents’ perceptions of their workplaces was originally developed by Dehghan et al. (2015) [15]. The HSSI scale comprises 18 items. These include 12-item observational questions and 6-item perceptual questions. These questions are related to workplace ambient (air temperature, humidity sensation, temperature of adjacent surfaces, air flow), perceived symptoms at work (intensity of thirst, sweating, and suffering from the heat, level of fatigue, heat stress symptoms), workplace settings (dimensions of the working space, types of working environment, ventilation system), working attire (types of clothes, colour, and material of working attire, personal protective equipment), the intensity of physical activity and body postures at work. Primary scores are allocated for each answer to every question, unlike the standardized four- or five-point Likert scale, commonly used [21]. The primary score of each question is multiplied by the effect coefficient to obtain the final score. The final scores will be added to become the total score and are then divided into three heat strain risk categories: (i) no or low heat strain (total score less than 13.5); (ii) likelihood of heat strain (total score between 13.6 to 18); and (iii) definitive heat strain (total score more than 18).
All the items in the original HSSI questionnaire have satisfactory factor loading of more than 0.40 (ranging from 0.44 to 0.90) and good internal consistency reliability with Cronbach’s alpha of more than 0.70 (ranging from 0.78 to 0.85) from its development process [15]. Even though the previous study in Malaysia utilized the HSSI questionnaire [18], no published Malay version of HSSI was recorded; hence, it would be more appropriate for the HSSI questionnaire to be translated into Malay, as most Malaysians can comprehend and communicate in the Malay language despite their different cultural backgrounds. The Malay version of the questionnaire would assist the respondents in sharing their actual experiences, judgments, and thoughts.
This study comprised two phases: phase 1 involves translating and adapting the original English version of HSSI into the Malay language based on the Guideline for Establishing Cultural Equivalency of Instruments [22] and the outline of the Questionnaire Development and Translation Process [23]. The researchers must understand the fundamental ideas, factors, and methodological issues involved, particularly in the quality of translation and the equivalence of results in different cultural and ethnic groups, because cross-cultural research has exceptional intricacies [24]. Phase 2 included Validation of the translated Malay version of HSSI, namely content validation, face validation, and psychometric analysis.
Translation and adaptations
Forward translation.
The initial translation from English to Malay was conducted by two independent translators who are well versed in both Malay and English and have background knowledge regarding Occupational and Environmental Health. This is to ensure the item and meaning of the translation resemble the original questionnaire. Each of the translators produced separate reports. The translated questionnaires from both translators were then compiled into a single translated questionnaire. Differences in word choice between the two translators were discussed until consensus was reached on which word to use.
Backward translation.
The finalized translated Malay questionnaire was then independently translated back to English by another two different translators with bilingual proficiency and background knowledge of Occupational and Environmental Health. Both these translators were not exposed to the original English Version of HSSI. This step is important to ensure the accuracy of the translation when it undergoes backward translation, as well as to identify any unclear or unsuitable wordings in the initial translation. Discrepancies between forward and backward translation were discussed between the researcher and all the translators. This is vital to ensure there were no differences in the wordings between the original items and the backward-translated version and assuring the translated questionnaire is appropriate for the Malaysian population, particularly the working community.
Validation of the translated Malay questionnaire
Content validation by expert committee.
Content validity refers to the extent to which a collection of items is considered together to establish an acceptable operational definition of a construct, mainly involving judgment or evaluation through careful conceptualization and domain analysis, especially in cross-cultural research [23, 24]. Methodological challenges include adapting the translated questionnaires into an understandable and culturally acceptable form among the local community while preserving the meaning of the original questionnaire [24].
Each committee member was given a thorough briefing on the content validation process, which was conducted using the widely used content validity index (CVI) [25]. The content validation form was emailed to all the experts, and they were given two weeks to complete the evaluation. Even though it would be more appropriate to conduct the content validation face-to-face approach, a non-face-to-face approach was chosen because of the COVID-19 pandemic, as well as to balance the cost and time challenges of gathering all the experts together [26]. The experts scored each item for relevance on a 4-point scale (1 = not relevant, 2 = somewhat relevant, 3 = quite relevant, and 4 = highly relevant).
Face validation (pilot-testing finalize draft of translated questionnaire).
Face validity is defined as the degree to which the raters evaluate the items of an instrument to be appropriate for the intended construct and objectives [26]. A face validation form comprising rating scores for clarity and comprehensiveness was developed to score individual items of the translated Malay version of HSSI. This ensures that the intended reviewers understand the task and evaluate accordingly. The ’clarity’ of commands, languages, and phrases applies to whether there were any inconsistencies or various ways to interpret the items, whereas the ’comprehensibility’ of commands, languages, and phrases indicates whether the words or sentences of the constructed items may well be easily understood by the reviewers [26]. It is of utmost importance to establish face or response process validity to support the overall validity of the translated Malay HSSI.
A small group of 10 reviewers was selected to rate and comment on the complete set of questionnaires. For face validation, a minimum of ten raters is required [26–29]. The reviewers included various groups of workers working in a hot environment or under the hot sun, including construction workers (two), landscape workers (three), outdoor cleaners (three), and rubbish collectors (two). Interviews based directly on the questionnaire were conducted among three illiterate workers. After reviewing and rating the questionnaire, the respondents were interviewed on their thoughts, understanding, and feedback on difficulties in answering the questionnaire, especially the suitability of the choice of words according to the community’s cultural context. This is to ensure the sampled respondents have a clear understanding, with no confusion, and that the translated HSSI items contained the same meaning or concept as the original questionnaire. These steps were considered crucial to confirm comprehensibility and to fulfill the face validity or response process validity before the pilot study.
Field study or psychometric analysis.
After the finalized draft questionnaire was ready, a cross-sectional field study was conducted from January 2022 to March 2022 among the working population, which shared similar characteristics and activities to the intended sample population. The respondents were different from the respondents from the pilot testing. The outdoor workers were selected through purposive sampling from various occupations exposed to sunlight or the hot environment during their working hours. They include construction workers, landscape workers, sweepers, cleaners, and rubbish collectors. The researcher approached them and invited them to participate in the field study. Workers who consented were then screened for eligibility based on the inclusion and exclusion criteria of the study. Workers who are more than 18 years old and exposed to heat or sun during their eight hours shift per day were included while workers who were pregnant and with underlying co-morbid like heart disease, autoimmune disorders, thyroid disease or mental illnesses were excluded. Self-administered questionnaires were given to the participants, and interviews were conducted strictly based on the questionnaire for illiterate workers. One hundred and eighty-eight outdoor workers were recruited for the field study. A minimum sample of 180 respondents was needed with a subject-to-item ratio of 10:1, as recommended by Hair et al. (2014) [30]. Returned questionnaires were examined for missing data or duplicated answers. Clarifications were made to questions that had more than two answers to ensure respondents chose the most appropriate answers to their understanding.
Construct validity. Construct validity refers to the extent to which the translated Malay version of HSSI was correlated to other measures based on hypothetically derived, pre-defined hypotheses [31]. The construct validity was conducted by performing the exploratory factor analysis (EFA).
Internal consistency. Internal consistency or reliability refers to how closely all test items assess the same notion or construct and is thus related to how closely the test items are related to one another. Before a test is used for research or examination purposes, internal consistency should be established to assure validity. Internal consistency defines the extent to which all the test items assess the same notion or construct and is thus related to how closely the test items are related to one another [32].
The intraclass correlation coefficient (ICC). ICC refers to the correlation within a class of variables or items for repeated measurements, and it was conducted to assess the test-retest reliability of the items [33]. Test-retest reliability is usually used to evaluate the degree of consistency to meet the objective of validating a questionnaire design, particularly during the initial pilot or field testing [34]. Data were collected from the participants using the same questionnaire three to four weeks after the first questionnaires were distributed.
Statistical analysis
Microsoft Excel was used for initial data collection and cleaned before data analysis was performed using IBM SPSS statistical package version 28.0. For descriptive statistics, normally distributed continuous variables were presented as mean and standard deviation, while categorical variables were presented as frequency and percentage. Both content validity index (CVI) and Face Validity Index (FVI) were computed using Microsoft Excel.
Content validity.
The content validity index (CVI) score was computed for both the item level (I-CVI) and the scale level (S-CVI) [35]. Before the calculation of I-CVI, the score of each item was re-coded from the 4-point relevance scale into two new categories in which items scored either 1 or 2 were allocated into the "not relevant" category with ’0’ point each while items scored either 3 or 4 were allocated into the "relevant" category with ’1’ point each. The accumulated points were then divided by the number of experts (six) to obtain the I-CVI value for each item.
The S-CVI is the average of the I-CVIs for all of the scale items. There are two approaches for calculating S-CVI: (i) calculation of the average of the I-CVI scores for all items on the scale (S-CVI/Ave); and (ii) the proportion of items on the scale that achieve universal agreement (a relevance scale of 3 or 4 by all experts or score ’1’ after re-coding) (S-CVI/UA) [35]. The definition and formula of the CVI indices are summarized in Table 1.
Face validity.
Face validity can be represented by the face validity index (FVI), which is executed and calculated based on the measures outlined by previous studies [26–29]. The degree of clarity and comprehensibility was rated according to the 4-point scale (1 –not clear and not understandable, 2 = somewhat clear and understandable, 3 = clear and understandable, and 4 = very clear and understandable).
Before the calculation of the face validity index (FVI), the score of each item was re-coded from the 4-point clarity and comprehensibility scale into 2 new categories in which items scored either 1 or 2 were allocated into the "not clear and understandable" category with ’0’ point each while items scored either 3 or 4 were allocated into the "clear and understandable" category with ’1’ point each. The FVI score was computed for both the item level (I-FVI) and the scale level (S-FVI) [26]. The accumulated points were then divided by the number of raters (10) to obtain the I-FVI value for each item.
The S-FVI is the average of the I-FVI scores for all of the scale items. There are two approaches for calculating S-FVI: (i) calculation of the average of the I-FVI scores for all items on the scale (S-FVI/Ave); and (ii) the proportion of items on the scale that achieve universal agreement (a clarity and comprehensibility scale of 3 or 4 by all reviewers or score of ’1’ after re-coding) (S-FVI/UA) [26]. The definition and formula of the FVI indices are summarized in Table 1.
Construct validity.
The construct validity was conducted by performing the exploratory factor analysis (EFA) using IBM SPSS Statistical Tool. The initial analysis included assessing the assumptions of factor analysis followed by generating and examining Bartlett’s test of sphericity, Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy, and communalities [36]. In order to determine the components that underlie each of the 18 items, factor extraction was carried out using principal axis factoring after the appropriateness of the initial analysis. The following analysis was used to determine the number of components to be retained: (i) eigenvalue larger than one rule; (ii) examining the scree plot test; and (iii) the need for at least 50% of the cumulative percentage of variance [30, 37]. The factors obtained were then rotated using varimax, an orthogonal rotation that attempts to minimize the number of variables that have high loadings on a factor and, thus, improves the factors’ interpretability [36, 37]. At least three measured items are required to statistically identify a component with loadings greater than 0.3 [38, 39].
Internal consistency.
Cronbach’s alpha was performed to approximate the internal consistency of each component or factor generated in PCA using IBM SPSS. The lowest accepted Cronbach’s alpha values were considered at 0.70, which verifies that the correlations among items on the same components are acceptable [32].
Interclass correlation coefficient (ICC).
ICC and their 95% confidence intervals were analysed in IBM SPSS Version 28.0, including mean-rating (k = 2), absolute agreement, and two-way mixed-effects model reliability [40]. Values between 0.75 and 0.9 and > 0.90 signify good and excellent reliability, respectively, based on the 95% confidence interval of the ICC estimate [40].
Ethical considerations
The original HSSI questionnaire with the calculation is shared by the researcher/developer in PDF format and set in the public domain with open access [15]. Additional permission from the author/developer of the original HSSI questionnaire was also obtained. Ethical approval was obtained from the Ethical Committee of the Faculty of Medicine and Health Sciences, Universiti Malaysia Sabah (UMS) [Approval Code: JKEtika 3/21 (18)]. The field study was conducted in accordance with the Declaration of Helsinki and any additional protocol or requirement from the UMS Ethical Committee. Informed consent was obtained from every participant prior to the study, either written consent which was signed by the participants or verbal consent which was witnessed by the workers’ supervisors or leaders. All the necessary information was explained to the respondents thoroughly, including the nature and purpose of the study, the procedures, and the tool used. The participants were assured that the collected data and information were strictly for research purposes and would not be used for any other purpose or disclosed to others without the participant’s permission. No personal information was revealed in the reports and publications. Participation in the study was entirely voluntary, with no payment given.
Results and discussion
Translation and adaptation
Forward translation.
The translated questionnaires from both translators were compiled into a single translated questionnaire. Differences in word choice between the two translators were discussed until consensus was reached on which suitable word to use. Amendments were made to ensure the words in the translated questionnaire were appropriate for the Malaysian population, particularly the outdoor working community.
Backward translation.
The finalized translated Malay questionnaire was then independently translated back to English by another two different translators with bilingual proficiency and background knowledge of Occupational and Environmental Health. Discrepancies between forward and backward translation were discussed between the researcher and all the translators. The meaning of most of the items which underwent backward translation was found to correspond to the original English questionnaire except item 1, item 4, item 8, and item 9. All four translators were sought for suggestions, and the researchers then worked on rephrasing the identified items.
- ■ In item 1, the translators decided to use the word ’panas’ instead of ’suam,’ which was the direct translation to ’warm’ as it is more understandable to describe the ambient temperature in the local context. Hence, the phrases’ slightly warm’, ’warm,’ and ’very warm’ were translated to ’sedikit panas,’ ’panas,’ and ’sangat panas’ to differentiate the different ambient temperatures.
- ■ In item 4, the ‘…pengaliran keluar masuk udara sejuk….’ was rephrased to ‘… peredaran udara sejuk…’ for the phrase ‘…cold weather circulation… ‘.
- ■ For item 8, the word ’intensiti’ was changed to ’tahap’ as ’intensiti’ is not commonly used in Malay, and the workers may have problems understanding the word.
- ■ For item 9, all the translators agreed to use the word ’terkesan’ instead of ’geram,’ which was deemed more appropriate among the local community than the original word ’annoyed.’
The preliminary version of the Malay version of HSSI was then finalized.
Content validation
An expert committee comprising six members comprising of two Occupational Health Doctors, an Environmental Health Officer, a Public Health physician, a Family Medicine Specialist, and a representative of the outdoor workers were consulted to review both versions of the translations and quantitatively evaluate the questionnaires.
All of them rated all items as 1.00 for the translated questionnaire. Hence the calculated I-CVI for all of the items and the S-CVI were 1.00. The relevance ratings on the item scale by the six experts and the calculation of different CVI indices for the Malay version of the HSSI questionnaire are illustrated in Table 2. This meant the content validation of the Malay version of HSSI was acceptable [35] and no item was further edited or removed.
Face validation
Face validity is defined as the degree to which the raters evaluate the items of an instrument to be appropriate for the intended construct and objectives [26]. Before tabulating the face validity, inter-rater reliability, which is to evaluate the consistency or agreement between the 10 reviewers/raters in their ratings were also conducted using intraclass correlation coefficient (ICC) with mean rating (k = 10), absolute-agreement, and two-way mixed effects model. The overall ICC value was 0.887 (95% CI; 0.789–0.951), which signifies good inter-rater reliability.
The ten reviewers’ clarity and comprehension ratings on the item scale and the calculation of different Face Validity Index (FVI) indices for the questionnaire, including translated Malay version of HSSI, are tabulated in Table 3. The calculated I-FVI for all of the items and the S-FVI were equal to or more than the minimum value of 0.83 [26, 28]. Hence, the face validation for the translated Malay version of HSSI is acceptable.
Psychometric analysis
Sociodemographic and work characteristics of participants.
A field study was conducted among the outdoor working population, which shared similar characteristics and activities to the intended sample population for the psychometric testing. Two hundred participants consented to be involved in the study, but 12 responses had to be excluded due to incompleteness. Hence, only 188 questionnaires were analyzed.
Most participants were male (65.4%) and were of Bajau ethnicity (22.3%). The age range of the respondents was from 17 to 62 (mean 35.7 ± 9.4) years old. 18.1% did not have formal education, while the remaining 81.9% had at least primary school level education. 75% of the participants earned a household income of less than RM2000 per month. Most participants were cleaners and rubbish collectors (23.9%), followed by sweepers (21.3%), landscape workers (18.1%), supervisors or guards (14.9%), and construction workers (14.4%). All the participants worked a minimum of eight hours a day, and 79.2% were exposed to sunlight or a hot environment for at least six hours per day. Most of the workers (44.1%) were usually walking, while 30.3% were standing with high mobility at work. The summary of the participants’ sociodemographic and work characteristics is tabulated in Table 4.
Exploratory factor analysis for construct validity.
The Kaiser-Meyer-Olkin measure for sampling adequacy was 0.812, which surpassed the recommended minimum value of 0.70 (Kaiser, 1974). Bartlett’s test of sphericity (test of the factorability of the correlation matrix) was also significant [χ2(153) = 1146.74, p<0.001], which indicates the items were correlated and the correlation matrix of the variables in the dataset diverges significantly from the identity matrix [30, 37]. The communality values, which assess how well each variable is explained by the factors are acceptable with values ranging from 0.415 to 0.804. Assumptions of multivariate normality and linearity were fulfilled. Multicollinearity is not present as Variance Inflation Factor (VIF) analysed using linear regression was < 10 (range:1.187 to 2.486). Thus, factor analysis was conducted using principal axis factoring with all 18 items. Factor loading was set at 0.30 to suppress all loadings less than 0.30.
Four factors were identified with eigenvalues of more than 1, which accounted for 56.32% of the cumulative percentage of variance. The first factor described 28.25% of the variance, the second factor explained 12.44% of the variance, while the third and fourth factors described 9.54% and 6.09% of the variance, respectively. After varimax rotation, all the components had at least three items with factor loadings greater than 0.40. Therefore, no item was removed. The factors extracted were labelled as:
- Factor 1: Physical activity and perceived symptoms at work
- Factor 2: Observation and Perception of workplace ambient
- Factor 3: Working attire and personal protective equipment
- Factor 4: Observation of workplace setting
Table 5 summarizes the results of the items, factor loadings, communalities, and eigenvalue for the rotated factors.
Cronbach’s alpha for internal consistency. Cronbach’s alpha was performed to approximate the internal consistency of the overall value and each factor. The overall Cronbach’s alpha was 0.776, while Cronbach’s alpha for each factor ranged from 0.705 to 0.758. Hence, the internal consistency reliability is acceptable and verifies that the correlations among items on the same factors were also acceptable [32].
The intraclass correlation coefficient (ICC) for test-retest reliability. The test-retest reliability was conducted on the total HSSI scoring using ICC with mean rating (k = 2), absolute-agreement, and two-way mixed effects model on 178 participants at 3–4 weeks intervals. Thirty-six participants were uncontactable and had to be dropped out from the retest. The remaining 152 participants were still a good sample size for test-retest reliability [40]. The overall ICC value was 0.792 (95% CI; 0.764–0.801), which signifies good reliability.
Discussion
This study executed the step-by-step measures suggested in the literature to develop a reliable and culturally-adapted Malay version of HSSI suitable for the working population in Malaysia. The Malay version of HSSI will be a good screening tool to assess the heat strain of workers exposed to sunlight or a hot environment at work before a precise and invasive assessment is conducted.
The forward and backward translations were conducted according to the measures recommended for cultural adaptation [22, 23, 41]. An expert committee which consists of various professionals related to occupational health as well as a representative of outdoor workers was consulted to review both versions of the translations as well as to evaluate the questionnaires quantitatively. Even though the team of developers of the original HSSI was not included in the expert committee, all of the identified discrepancies and inconsistencies were discussed, and amendments were conducted without affecting the meaning of the original items. The calculated I-CVI for all of the reviewed items and the S-CVI were 1.00 after item enhancements based on the experts’ recommendations and were reviewed for the second time by all the experts. Furthermore, the calculated I-FVI for all of the items and the S-FVI were equal to or more than the minimum value of 0.83 [26, 28]. These indicated that both the content and face validity of the Malay version of HSSI is acceptable.
Factor analyses using principal component analysis with Varimax rotation were conducted to extract the main factors of the 18-item HSSI. This procedure was chosen to evaluate the construct validity of the translated questionnaire by exploring the correlations among the items in the Malay version of HSSI to identify the variables’ underlying structure. No item was omitted during the translation process and psychometric testing.
There were only three main domains in the original HSSI, which were labelled as subjective judgment, protective clothes, and physical activities [42]. However, this study extracted four domains or factors from the PCA for the Malay version of HSSI. They were labelled as (i) Physical activity and perceived symptoms at work (seven items); (ii) Perception of workplace ambient (four items); (iii) Working attire and personal protective equipment (four items); and (iv) Personal observation of the workplace setting (three items). All of the factors extracted had an acceptable level of internal consistency reliability, with each factor having Cronbach’s alpha value of more than 0.70 [32]. The test-retest reliability of the HSSI scoring was also acceptable, with a good ICC of 0.792.
Item 6, 7, 8, 9, and 18 are loaded together with Item 5 and 17, which point to different understandings of these items in the Malaysian setting. Physical activity was combined with perceived symptoms at work. One probable explanation for this combination is the intensity of physical activity at work is connected and has effects on the symptoms experienced by the workers. Furthermore, items 10,11 and 12 were also loaded separately, and the new factor labelled ’Observation of the workplace setting’ was created for these items. One likely explanation was that most workers (86.2%) worked outdoors, and their interpretations were towards their outdoor working environments.
Study limitations
There were a few limitations identified in the study. Even though the sample size of the field study fulfills the criteria of a minimum subject-to-item ratio of 10:1, as recommended by Hair et al. (2014) [30], it would be more appropriate to have a larger sample size to diminish error in data and ensure reliability for the factors extracted. Two-thirds of the workers were male workers who may not perceive heat as an occupational hazard as they were used to working long hours outdoors.
Structured interviews were conducted among participants who were illiterate or with a low level of education which was assisted by translators who were well versed in the local language (Bajau and Kadazan/Dusun). Therefore, social desirability bias could be considered as the workers may tend to give responses based on what was more appealing to them, which could be overrated or underrated, and the responses given were more favourable to the researcher rather than their actual experiences.
Most of the workers in the field study mainly worked outdoors, while participants who were supervising or guarding were working indoors and outdoors. No worker was specifically working in the hot indoor environment. Therefore, there may be a limitation in using this Malay version of questionnaire among those working in the hot indoor environment. Besides, the field study was mainly conducted among the outdoor workers in the Borneo state of Sabah and northern Peninsular Malaysia. Hence, further studies among Malay-speaking workers, especially those working in hot indoor environments, are crucial to validate the findings of this study.
Conclusion
The findings from this study indicate that the Malay version of the Heat Strain Score Index is a valid and reliable instrument. Confirmatory factor analysis (CFA) will be need to be conducted once data is available from the actual study population to further add value to the construct validity, convergent validity, discriminant validity and composite reliability of the translated questionnaire. It can therefore be used to assess the heat strain among the Malay-speaking susceptible workers in Malaysia who are exposed to hot, humid environments. This ensures that further assessment and adequate control and prevention measures can be taken based on the level of heat strain identified.
Acknowledgments
The researchers are grateful for the commitment from all the translators, experts, and reviewers involved in the initial translation phase, content validation, and face validation to enable the finalized draft of the Malay version of the HSSI questionnaire. Besides, the researchers were also appreciative of all the workers who agreed to participate in the field study and gave their total cooperation throughout this study.
References
- 1. Parsons K. Human Thermal Environments: The Effects of Hot, Moderate, and Cold Environments on Human Health, Comfort, and Performance, Third Edition. (2014). p. 1–586.
- 2. Environmental Protection Agency (EPA). (2016). Climate Change and the Health of Occupational Groups. Retrieved from https://www.cmu.edu/steinbrenner/EPA%20Factsheets/occupational-health-climate-change.pdf on 11 April 2022.
- 3. Varghese B. M., Hansen A., Bi P., & Pisaniello D. (2018). Are workers at risk of occupational injuries due to heat exposure? A comprehensive literature review. Safety Science, 110(May), 380–392. https://doi.org/10.1016/j.ssci.2018.04.027
- 4. Xiang J., Bi P., Pisaniello D., & Hansen A. (2014). Health impacts of workplace heat exposure: An epidemiological review. Industrial Health, 52(2), 91–101. pmid:24366537
- 5.
Jacklitsch B., Williams W., Musolin K., Coca A., Kim J.-H., & Turner N. (2016). NIOSH criteria for a recommended standard: occupational exposure to heat and hot environments. US Department of Health and Human Services, Publication 2016–106.
- 6. Sim L.L. (2018). DOSH: Workers at risk of heat stress. The Star. Retrieved from https://www.thestar.com.my/news/nation/2018/08/26/dosh-workers-at-risk-of-heat-stress-awareness-needed-to-minimise-exposure on 30 May 2022
- 7. Zander K. K., & Mathew S. (2019). Estimating economic losses from perceived heat stress in urban Malaysia. Ecological Economics, 159(June 2018), 84–90. https://doi.org/10.1016/j.ecolecon.2019.01.023
- 8. Department of Occupational Safety and Health (DOSH). (1994). Occupational Safety and Health Act (Act 514) 1994. Retrieved from https://www.dosh.gov.my/index.php/legislation/acts-legislation/23-02-occupational-safety-and-health-act-1994-act-514/file on 12 June 2022
- 9. Roghanchi P., & Kocsis K. C. (2018). Challenges in Selecting an Appropriate Heat Stress Index to Protect Workers in Hot and Humid Underground Mines. Safety and Health at Work, 9(1), 10–16. pmid:30363068
- 10.
Roghanchi P., Sunkpal M., & Kocsis K. C. (2015). Understanding the Human Thermal Balance and Heat Stress Indices as They Apply to Deep and Hot US Mines Understanding the Human Thermal Balance and Heat Stress Indices as they apply to Deep and Hot US Mines. 15th North America Mine Ventilation Symposium, August, 0–7.
- 11. Brake R., & Bates G. (2002). A valid method for comparing rational and empirical heat stress indices. Annals of Occupational Hygiene, 46(2), 165–174. pmid:12074026
- 12. Epstein Y., & Moran D. S. (2006). Thermal comfort and the heat stress indices. Industrial Health, 44(3), 388–398. pmid:16922182
- 13. Lazaro P., & Momayez M. (2020). Development of a modified predicted heat strain model for hot work environments. International Journal of Mining Science and Technology, 30(4), 477–481. https://doi.org/10.1016/j.ijmst.2020.05.009
- 14. Tikuisis P., McLellan T. M., & Selkirk G. (2002). Perceptual versus physiological heat strain during exercise-heat stress. Medicine and Science in Sports and Exercise, 34(9), 1454–1461. pmid:12218738
- 15.
Dehghan H., Bagher Mortzavi S., Javad Jafari M., & Reza Maracy M. (2015). Development and Validation of a Questionnaire for Preliminary Assessment of Heat Stress at Workplace ARTICLE INFORMATION ABSTRACT. In JRHS Journal of Research in Health Sciences J Res Health Sci (Vol. 15, Issue 3). www.umsha.ac.ir/jrhs
- 16. Dehghan H., & Mortazavi S., & Jafari M., & Meraci M., & Khavanin A., & Jahangiri M. (2011). Construct Validation of A Heat Strain Score Index With Structural Equation Modeling. Health System Research, 6(4), 601–612. https://www.sid.ir/en/journal/ViewPaper.aspx?id=208666
- 17. Bolghanabadi S., Ganjali A., & Ghalehaskar S. (2019). Investigation of thermal exposure in traditional neyshabur bakeries using heat strain and physiological indices. MethodsX, 6(February), 355–359. pmid:30847283
- 18. Yatim S. R., Rajan S., Shafie F. A., Che Dom N., Zainudin N. S., & Ishak A. R. (2019). Heat stress and potential of heat strain among solid waste collectors in Hulu Selangor Local District. Health Scope, 1, 193–197.
- 19. Jafari M. J., Teimori G., Khodakarim S., & Assilian-Mahabadi H. 2015. Investigation the relationship between heat strain score index and physiological parameters among open pit miners. Safety promotion and injury prevention (Tehran), 3(4), 213–222.
- 20. Alshammary N. F., Salama K. F., & Zafar M. (2020). Heat Stress among Steel Workers in Al Jubail,Saudi Arabia. Journal of Ecophysiology and Occupational Health, 20(1 & 2), 34–40. https://doi.org/10.18311/jeoh/2020/24661
- 21.
Nemoto, T., & Beglar, D. (2014). Developing Likert-Scale Questionnaires. JALT 2013 Conference Proceedings, 1–8.
- 22.
Ohrbach R., Bjorner J., Jezewski M., John M. T., & Lobbezoo F. (2009). Guidelines for establishing cultural equivalency of instruments. New York: University at Buffalo.
- 23. Tsang S., Royse C. F., & Terkawi A. S. (2017a). Guidelines for developing, translating, and validating a questionnaire in perioperative and pain medicine. Saudi Journal of Anaesthesia, 11(5), S80–S89. pmid:28616007
- 24. Sperber A. D. (2004). Translation and Validation of Study Instruments for Cross-Cultural Research. Gastroenterology, 126(1). pmid:14978648
- 25. Davis L. L. (1992). Instrument Review: Getting the Most From a Panel of Experts.
- 26. Yusoff M. S. B. (2019). ABC of Response Process Validation and Face Validity Index Calculation. Education in Medicine Journal, 11(3), 55–61. https://doi.org/10.21315/eimj2019.11.3.6
- 27. Lau A. S. Y., Yusoff M. S. B., Lee Y. Y., Choi S. B., Xiao J. Z., & Liong M. T. (2018). Development and Validation of a Chinese translated questionnaire: A single simultaneous tool for assessing gastrointestinal and upper respiratory tract related illnesses in pre-school children. Journal of Taibah University Medical Sciences, 13(2), 135–141. https://doi.org/10.1016/j.jtumed.2017.11.003
- 28. Marzuki M. F. M., Yaacob N. A., & Yaacob N. M. (2018). Translation, Cross-cultural Adaptation and Validation of System Usability Scale (Malay Version) Questionnaire for the Assessment of Mobile Application. JMIR Human Factors, 5(2).
- 29. Ozair M. M., Baharuddin K. A., Mohamed S. A., Esa W., & Yusoff M. S. B. (2017). Development and Validation of the Knowledge and Clinical Reasoning of Acute Asthma Management in Emergency Department (K-CRAMED). Education in Medicine Journal, 9(2), 1–17. https://doi.org/10.21315/eimj2017.9.2.1
- 30.
Hair J. F., Black W. C., Babin B. J., Anderson R. E., & Tatham R. L. (2014). Pearson new international edition. Multivariate data analysis, Seventh Edition. Pearson Education Limited Harlow, Essex.
- 31. te Molder M. E. M., Vriezekolk J. E., Bénard M. R., & Heesterbeek P. J. C. (2021). Translation, cross-cultural adaptation, reliability and construct validity of the Dutch Oxford Knee Score–Activity and Participation Questionnaire. BMC Musculoskeletal Disorders, 22(1). pmid:34404359
- 32. Tavakol M., & Dennick R. (2011). Making sense of Cronbach’s alpha. In International journal of medical education (Vol. 2, pp. 53–55). pmid:28029643
- 33. Liljequist D., Elfving B., & Roaldsen K. S. (2019). Intraclass correlation–A discussion and demonstration of basic features. PLoS ONE, 14(7). pmid:31329615
- 34. Bujang M. A., & Baharum N. (2017). A simplified guide to determination of sample size requirements for estimating the value of intraclass correlation coefficient: a review. In Archives of Orofacial Sciences The Journal of the School of Dental Sciences, USM Arch Orofac Sci (Vol. 12, Issue 1).
- 35. Polit D. F., & Beck C. T. (2006). The content validity index: Are you sure you know what’s being reported? Critique and recommendations. Research in Nursing and Health, 29(5), 489–497. pmid:16977646
- 36. Watkins M. W. (2018). Exploratory factor analysis: A guide to best practice. Journal of Black Psychology, 44(3), 219–246.
- 37. Williams B., Onsman A., Brown T., Andrys Onsman P., & Ted Brown P. (2010). Exploratory factor analysis: A five-step guide for novices. Journal of Emergency Primary Health Care (JEPHC), 8(3).
- 38.
Samuels P. (2016). Advice on exploratory factor analysis. Centre for Academic Success, Birmingham City University, UK. https://doi.org/10.13140/RG.2.1.5013.9766
- 39. Izquierdo Alfaro I., Olea Díaz J., & Abad García F. J. (2014). Exploratory factor analysis in validation studies: Uses and recommendations. Psicothema.
- 40. Koo T. K., & Li M. Y. (2016). A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. Journal of Chiropractic Medicine, 15(2), 155–163. pmid:27330520
- 41. Sousa V. D., & Rojjanasrirat W. (2011). Translation, adaptation and Validation of instruments or scales for use in cross-cultural health care research: A clear and user-friendly guideline. In Journal of Evaluation in Clinical Practice (Vol. 17, Issue 2, pp. 268–274). pmid:20874835
- 42. Dehghan H., Mortazavi S. B., Jafari M. J., Meraci M. R., Khavanin A. L. I., & Jahangiri M. (2011). Construct Validation of a heat strain score index with structural equation modeling.