Abstract
BACKGROUND:
The Quick Exposure Check (QEC) is an instrument to identify occupational risk factors, and to screen subjects for the risk of musculoskeletal disorders in their workplace.
OBJECTIVES:
The aim of this prospective observational study was the cross-cultural adaptation of the QEC for Persian (QEC-Pr) speaking workers and the determination of the minimally required psychometric properties, including reliability and validity, in a Persian population.
METHODS:
The forward-backward translation was completed according to the published guidelines. A total of 217 male construction workers were recruited. The Rapid Upper Limb Assessment (RULA) tool was used as the criterion. The Inter- and intra-rater reliability was performed respectively in a sub-sample of n = 50 and n = 30. Convergent validity was determined from the correlation between the QEC-Pr and the RULA. Internal consistency was assessed using Cronbach’s α.
RESULTS:
High levels of inter-rater reliability (ICC2.1 range = 0.79–0.93), intra-rater reliability (ICC2.1 range = 0.74–0.89) and internal consistency (α= 0.74) were obtained. Concurrent validity between the different sections of the QEC and RULA scores were determined to be fair but below the desired level (range r = 0.41–0.44).
CONCLUSIONS:
The QEC-Pr can be applied as a specific risk assessment instrument for different occupational situations, such as the construction industry, in Persian language populations.
Keywords
Introduction
Work-related musculoskeletal disorders (MSDs) are a common health problem in modern societies [1–3] with substantial health costs and impacts on quality of life and productivity [2]. Between 2009 and 2011, there was an estimated US$213 billion in annual cost for direct treatment for and lost wages due to MSDs in the United States [4]. In the Netherlands, it is estimated that work-related repetitive strain injury (RSI) costs a total of US$2.1 billion each year [5]. Studies of Swedish back and neck pain patients on sick leave have estimated the annual costs about 3.5 billion Euros, corresponding to nearly one-third of the nation’s total health-care costs [6, 7]. The most involved body regions are the low back (48%), upper extremities (18%), knee (5%) and ankle/foot (2%) plus multiple sites (5%) [8]. The MSDs also present a major occupational health problem in Iran, however the overall prevalence reported indicates a difference in affected regions being predominantly in the upper limb (54.2%), with the back (50%) and neck (31.8%) [9, 10]. As with most chronic diseases, MSDs have multiple risk factors including physical, psychosocial and individual factors [4] as contributed exposure variables that can be present independently or interactively [11], although it is recognized that the combined effects of these risk factors are greater than the sum of the individual contributors [12].
Due to the high prevalence and incidence of MSDs, accurate identification, evaluation and measurement of these risks can assist in conducting risk assessment programs and in particular in ergonomic research [10]. Such assessment instruments for MSDs risk factor exposure are considered in three formats: patient reported outcome measures (PROMs), observational instruments and psycho-physiological methods [13].
The PROMs include: the Numeric Rating Scale (NRS), in preference to the Visual Analogue Scale (VAS) [14], and the Nordic Questionnaire [13], which collects data through either an interview or direct patient responses. Their advantages include the simplicity of use, low user and clinician burden, low cost and their application in surveying a large numbers of subjects. The main disadvantages include a reduced precision and reliability in the perceptions of worker exposure [15].
The observational tools include the Quick Ex-posure Check (QEC), the Rapid Upper Limb Assessment (RULA) tool and the Rapid Entire Body Assessment (REBA), with advantages that favor an observer being able to rapidly and effectively assess and record workplace exposure factors [16, 17] in a way that is inexpensive and simple to complete. Their disadvantages include the presence of low inter- and intra-rater (test-retest) reliability [18]. The psycho-physiological methods record data and measure the exposure factors via sensors attached to the body. These instruments include electro-goniometers, electromyography and inclinometers. Their advantages are the high practicality in research settings and the reliability and precision of the instruments. Their disadvantages are the cost of purchase, maintenance and a lack of practicality in the industrial settings [19].
The choice of an appropriate measurement technique should be based on the requirements of the measurement strategy and the target population. Many exposure assessment methods are either observa-tional or self-report [20]. However, it is recommended to employ at least two methods concurrently to achieve a comprehensive and scientifically sound exposure assessment [21].
The QEC [22, 23] provides the observer with a flexible technique that can be applied in various occupational environments and conditions. It is applicable to both static and dynamic occupations, requires around 10-15 minutes of training for professional competency and can be completed in around 10 minutes for each task [1, 13]. A participatory approach is emphasized in its completion whereby workers themselves play a crucial role in the risk exposure identification [24]. The QEC is shown to be both valid and reliable in English [17], Chinese [24], Brazilian-Portuguese [25] and Turkish [1]. The original QEC has been used in several Iranian studies however its psychometric characteristics are yet to be satisfactorily evaluated in Persian in a suitable culturally adapted and translated format.
Consequently, the aim of this study was the cross-cultural adaptation of the QEC to Persian (QEC-Pr) and the determination of its psychometric characteristics in a Persian working population, including reliability, validity, standard error of measurement (SEM) and internal consistency. This will enable the QEC-Pr to be subsequently compared with the original English and other language versions as well as being used in multicultural geographical areas and to facilitate data pooling and meta-analysis.
Methods
Measures/questionnaires
Quick Exposure Check (QEC)
The QEC is a risk exposure measurement tool that includes physical, organizational and psychosocial factors. It was developed by Li and Buckle in 1998 [26] and modified by David et al. in 2003 [23]. The tool is used by an observer to assess the postures and movements of four main body areas; the back, wrist/arms, shoulder/hand and the neck. The observed worker provides input regarding the amount of weight handled, time to complete the task, level of hand force exertion, visual demands, vibration, driving a vehicle, work pace and the level of stress [22]. The total score is calculated from the observer and worker scores with a range from 46-269 points with the risk profile categorized as: low (46-84 points), moderate (106-138 points), high (168-198 points) and very high (187-242 points) [17].
Rapid Upper Limb Assessment (RULA)
The RULA is a single page worksheet developed by McAtamney and Corlett of the University of Nottingham’s Institute of Occupational Ergonomics in 1993 [16]. The tool has a summated total ranging between one and seven that is assumed from four sub-scores of A through D: A groups the upper arm, lower arm and wrist postures; B groups the neck, trunk and legs postures; C = Score A + muscle use and force scores; D = Score B + muscle use and force scores [27–29].
We selected the RULA in this study to assess concurrent validity because the main features categorize the body postures and force, with action levels for assessment similar to the QEC which evaluates the exposure of individual workers to ergonomic risk factors including body posture, force exertion, and repetition [19]. Further, the RULA highlights the upper limb which is the region of highest problem prevalence in Iran [9, 10]. The RULA method also provides an overall score that takes into account postural loading on the whole body, with particular attention to the neck, trunk, shoulders, arms and wrists. The overall score also considers the time the posture is held, the force used and the repetitiveness of the movement [26, 29]. It seems that the RULA, as with the QEC, evaluates risk factors of body posture, force exertion and repetition.
Translation and cross-cultural adaptation
The cross-cultural adaptation and translation of the English version of the QEC into Persian was conducted according to published guidelines [30, 31]. Forward translation was performed by two independent native Persian speakers. One translator was an ergonomist and aware of the check concept and the other was not. After discussing discrepancies between each translator’s versions a consensus version was adopted. Two independent and blinded translators subsequently performed a backward translation [32]. An expert review committee consisting of one physical therapist, one occupational health profession, one ergonomist, one psychometrician, all of the translators and the principal author, produced a pre-final consensus draft version of the QEC-Pr.
Face validity assessment
A total of 27 participants (12 experts and 15 workers) completed the pre-final QEC-Pr in order to test the alternative wording and to check readability, comprehension, interpretation and cultural relevance of the translation. They reported that the QEC was easy to understand and relevant, which established the version for psychometric testing in a sample of Persian construction workers.
Participants
A convenience sample of male construction workers from the labor trades was recruited. In Iran the construction industry workers are almost exclusively male due to cultural reasons and the physical and work environment requirements. This study was conducted in two phases of reliability and validity assessment. The reliability phase was conducted from the perspectives of both the workers and observers and the combination of answers of the observer and the workers were considered, as proposed by the QEC. We tested the inter- and intra-rater reliability. For the inter-rater reliability phase, two occupational health experts conducted the assessment on n = 50 construction workers. For the intra-rater reliability, one expert (SA) used a sample of n = 30 workers performing the same task to complete the test on two different days (3–10 days interval).
For the validity phase, 217 native Persian speaking construction workers from the labor trades were recruited. The inclusion criteria were: a) more than one year’s experience in the current occupation; b) good Persian language comprehension; c) aged 18–45 years; and d) written consent to participate and agreement to follow the research protocol. Exclusion included subjects with any infectious disease or documented condition that may affect working capacity and participation such as malignancy or depression.
In order to complete the checklist, each worker was observed simultaneously by the two trained experts for a period of 10-15 minutes while performing a specific task. The RULA and the first section of the QEC were completed by the two experts. The second section of the QEC was completed via an interview with the worker by each expert separately. These procedures were repeated only for the QEC by one of the experts (SA) on 30 workers within a 3-10 day interval in order to assess intra rater reliability.
The ethics committee of the University of Social Welfare and Rehabilitation Sciences (USWR) ap-proved the study (IR.USWRREC.1398016).
Statistics
Test-retest reliability (relative reliability) was assessed through Intraclass correlation coefficients and absolute reliability through SEM. Two types of test-retest reliability included the inter-rater and intra-rater both using the Intra-class Correlation Coefficient type 2,1 (ICC2,1) in randomly selected sub-samples, n = 50 for inter-rater and n = 30 for intra-rater. An ICC2.1 ≥ 0.70 was considered acceptable [33].
The SEM was calculated using the formula of [SD
Internal consistency was assessed using Cron-bach’s-α, with values 0.70–0.95 considered acceptable as the level is respectively high and indicates no presenting item redundancy [36, 37].
The concurrent validity was evaluated from calculating Pearson’s correlation r between the QEC-Pr and the RULA scores, where a minimum correlation of r = 0.70 is high and acceptable, while 0.60-0.70 is moderate and 0.40-0.60 is fair but both are below the desired level [37, 38].
All statistical analyses were calculated using the Statistical Package for the Social Sciences version 16 (SPSS 16) for Windows. The level of significance was set at p < 0.05.
Results
Participants’ characteristics
A total of 217 male construction workers (mean age = 31.7±5.98 years) for evaluating the validity, and sub-samples of 50 for the inter-rater and 30 for the intra-rater reliability were randomly selected from a convenience sample to participate in the reliability analysis. Demographic characteristics of the study sample are reported in Table 1.
Socio-demographic characteristics of the subjects (n = 217) and reliability
Socio-demographic characteristics of the subjects (n = 217) and reliability
During translation (forward and backward), no major difficulties were reported in completing the QEC. Some minor modifications were applied to the text based upon cultural relevance. In the first column, item B, “static” was translated to “without motion” and in the second column item L, “visual demand” was changed to “visual requirement”. Further, there was no missing data and all items were addressed.
Reliability
The results of the reliability analysis are shown in Tables 2 and 3. The ICC2.1 range values were high for both inter-rater (r = 0.79-0.93) and intra-rater (r = 0.74-0.89), indicating excellent reliability.
Reliability results (inter-rater) for the items and total score of the QEC
Reliability results (inter-rater) for the items and total score of the QEC
SEM = Standard error of measurement, SD = Standard deviation, ICC2.1 = Intraclass correlation coefficient.
Reliability results (intra-rater) for the items and total score of the QEC
Cronbach’s alpha = 0.74, which indicated a high level of internal consistency.
Concurrent validity
Concurrent validity was acceptable between: the sum of the score of the shoulder/arm and wrist/hand sections of the QEC with the C score of the RULA (r = 0.44); the sum of the score of the back and neck of the QEC with the D score of the RULA (r = 0.41); and the total score of the QEC with the Total RULA score (r = 0.43).
Discussion
This study was conducted in a sample of Persian construction workers in order to translate and cross-culturally adapt the original QEC into Persian and also evaluate its psychometrics properties. The QEC is a simple, user friendly and easily completed risk assessment instrument and consequently preferred to other risk assessment tools. Another QEC advantage, in contrast to most observational methods, is its combination of the workers’ answers with the observer’s perspectives simultaneously in the assessment [31]. In order to use an instrument in other languages the original conceptual content should be maintained through an appropriate cultural adaptation and linguistic translation [39, 40]. During this process no major difficulty or lack of clarity was found by workers or observers and none required help or added description to complete the QEC. Minor changes were made in the back section of the observer assessment: “select worse case situation” was translated to “select the worst body posture” and in the worker assessment column section L, “visual demand” was changed to “visual need” for cultural reasons as the concept of “demand” is unfamiliar in Persian society.
We considered the inter-and intra-rater reliability, internal consistency and measurement error within the reliability domain. Our results showed high inter-and intra-rater reliability comparable to previous versions.
Inter-rater reliability (ICC2.1range r = 0.79-0.93) was high for each section and the total score. Our result was very close to the Chinese version (0.71-0.97) [24] and higher than the Brazilian version (0.62–0.86) [31]. In the Turkish version, inter-rater reliability was not reported [1].
The intra-rater (test-retest) reliability was high (ICC2.1 ranger = 0.74-0.89) which was close to the findings in the Turkish version (0.60-0.84) [1] and higher than the Brazilian version findings (0.41-0.60) [31]. Test-retest reliability of the workers assessment section in the Chinese version was evaluated using Kappa values which fell within the range 0.68-1.00, demonstrating substantial agreement among workers on the exposure factors between the two assessment occasions. A reliability finding comparable to that found in this study.
The difference in our study finding compared to the Brazilian version showed the inter-rater reliability was higher than the intra-rater reliability. This may be related to the experience and expertise of our two observers who assessed the intra-reliability tests. It is anticipated that the reliability of a tool will increase as practitioners gain more experience in its use in the workplace. Two experts in health occupational engineering with over four years work experience conducted the inter-rater reliability which can ensure a low variability in completing the QEC.
When our results are compared with the four previous versions [1, 31], we found a higher intra-rater (test-retest) reliability. This may be ex-plained by the high education level of our participants and the greater experience of our observers. Another explanation may be related to the number of the tasks evaluated in our study compared to others. We evaluated only a single task for inter- and intra-rater reliability, which can ensure less variability and result in higher reliability. Furthermore, the internal consistency of the QEC was evaluated as in the Brazilian version (α= 0.76) [31] with a very similar result (α= 0.74). Evaluating the concurrent validity with the C, D and total score of the RULA showed a fair correlation range (r = 0.41-0.44). In the Brazilian version, concurrent validity was assessed using the Job Factors Questionnaire with similar results of a fair validity (r = 0.38) [31]. However the results still confirm previous studies findings that the level of concurrent validity is not optimal being at r = 0.41-0.44; below the desired 0.70 level [33]. We used the RULA method as, like the QEC, it assesses exposure factors of posture (upper body and limb), load/force and movement frequency [19]. In addition, we had no other valid and reliable instrument for use in Persian.
Strengths and limitations
The strengths of this study were the use of Persian speaking participants in their home geographical region. Persian is spoken by over 150 million people and is among the top 20 geographically spoken languages globally [41]. The inclusion criteria ensured all participants had an existing upper limb condition and that they found the translated version both linguistically and culturally acceptable. This study has provided a Persian version of the QEC which is the only available risk exposure observational outcome measure that is culturally adapted in Persian and is applicable to the Persian speaking populations.
One of the limitations of this study was the inability to use direct measurement methods for comparison with the QEC. Further, generalizability of the results is limited as the subjects were selected from the construction industry and not the general working population. Consequently there is no female gender representation balance as the construction industry in Iran is almost exclusively male workers due to cultural practice and the physical demands of the tasks: including manual handling; the physicality and work demands; the working hours; the work facilities for hygiene; and the physical proximity of workers to each other during both task and break time. We had some limitations in the selection of subjects available under the same conditions at two different test times that resulted in this test-retest reliability component being limited to 30 participants. Intra-rater reliability should be done in a constant time and in stable conditions; however the tasks of the participants changed over the time period of the study.
Conclusion
To the best of our knowledge, this developed Persian version of the QEC is the only risk exposure outcome measure available for Persian speaking workers. The QEC-Pr demonstrated satisfactory reliability and fair validity which were consistent with the characteristics of the instrument found in previous translation and cultural adaptation studies. This instrument can be used by occupational health researchers and ergonomists to measure occupational risk factors in the workplace of Persian language workers. However, the QEC-Pr requires further research, revision and refinement based on its application in the workplace to increase its generalizability to both genders and different work and occupational settings.
Conflict of interest
None to report.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Footnotes
Acknowledgments
The authors are grateful to the volunteers for participating in this study and to Mr Javad Nouri, Occupational Health Engineer, for his kind help in a section of the data gathering.
