Applied Ergonomics 45 (2014) 1257e1262
Contents lists available at ScienceDirect
Applied Ergonomics journal homepage: www.elsevier.com/locate/apergo
Validity and inter-observer reliability of subjective hand-arm vibration assessments Pieter Coenen a, Margriet Formanoy b, *, Marjolein Douwes b, Tim Bosch b, Heleen de Kraker b a Research Institute MOVE, Faculty of Human Movement Sciences, VU University Amsterdam, Van der Boechorststraat 9, 1081BT Amsterdam, The Netherlands b TNO Sustainable Productivity and Employability, Polarisavenue 151, 2132 JJ Hoofddorp, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 March 2013 Accepted 15 March 2014
Exposure to mechanical vibrations at work (e.g., due to handling powered tools) is a potential occupational risk as it may cause upper extremity complaints. However, reliable and valid assessment methods for vibration exposure at work are lacking. Measuring hand-arm vibration objectively is often difficult and expensive, while often used information provided by manufacturers lacks detail. Therefore, a subjective hand-arm vibration assessment method was tested on validity and inter-observer reliability. In an experimental protocol, sixteen tasks handling powered tools were executed by two workers. Hand-arm vibration was assessed subjectively by 16 observers according to the proposed subjective assessment method. As a gold standard reference, hand-arm vibration was measured objectively using a vibration measurement device. Weighted k’s were calculated to assess validity, intra-class-correlation coefficients (ICCs) were calculated to assess inter-observer reliability. Inter-observer reliability of the subjective assessments depicting the agreement among observers can be expressed by an ICC of 0.708 (0.511e0.873). The validity of the subjective assessments as compared to the gold-standard reference can be expressed by a weighted k of 0.535 (0.285e0.785). Besides, the percentage of exact agreement of the subjective assessment compared to the objective measurement was relatively low (i.e., 52% of all tasks). This study shows that subjectively assessed hand-arm vibrations are fairly reliable among observers and moderately valid. This assessment method is a first attempt to use subjective risk assessments of hand-arm vibration. Although, this assessment method can benefit from some future improvement, it can be of use in future studies and in field-based ergonomic assessments. Ó 2014 Elsevier Ltd and The Ergonomics Society. All rights reserved.
Keywords: Hand-arm vibration Risk assessment Validity
1. Introduction Exposure to mechanical vibrations at the workplace, such as hand-transmitted vibrations, can arise in numerous labors (e.g., in construction or manufacturing industries), for example when manually handling powered tools. These hand-transmitted vibrations are associated with a variety of signs and symptoms including vascular and neurological disorders (Griffin and Bovenzi, 2002). More specifically, it has been shown in several reviews that handtransmitted vibrations are associated with upper extremity
Abbreviations: HARM, hand-arm risk assessment method; ICC, Intra-class correlation coefficient. * Corresponding author. Tel.: þ31 88 866 52 90. E-mail addresses: [email protected] (P. Coenen), [email protected] (M. Formanoy), [email protected] (M. Douwes), [email protected] (T. Bosch), [email protected] (H. de Kraker). http://dx.doi.org/10.1016/j.apergo.2014.03.003 0003-6870/Ó 2014 Elsevier Ltd and The Ergonomics Society. All rights reserved.
complaints (Hagberg, 2002; Kittusamy and Buchholz, 2004; Punnett, 2004); for example, shoulder pain (van der Windt et al., 2000) and specific pathologies like tenosynovitis and epicondylitis (Palmer et al., 2007; Shiri and Viikari-Juntura, 2011). Although evidence is slightly inconsistent as there are also studies reporting weak evidence for the association of hand-arm vibration and upper-extremity complaints (da Costa and Vieira, 2010; Schweigert, 2002; van Rijn et al., 2010), it is generally accepted that hand-arm vibrations are an occupational risk. As a result of these potential occupational hazards, in 2002, European directives were communicated providing workers’ exposure limits for whole-body and hand-arm vibrations (2002/44/ EC). These directives, that are based on health and safety requirements, specify the maximum intensity of vibrations a worker can be exposed to, considering the duration of this specific vibration. Assessment of hand-arm vibrations is therefore based on both the duration and the intensity of the exposure. This approach is
1258
P. Coenen et al. / Applied Ergonomics 45 (2014) 1257e1262
supported by several studies showing that vibration, in which the exposure is a multiplication of duration and intensity of the handarm vibration, is associated with upper extremity disorders (Bovenzi, 2012; Griffin, 2004; Sauni et al., 2009). Therefore, duration as well as intensity of vibration should be considered to quantify the potential risk of musculoskeletal disorders of handarm vibrations. Objectively measuring hand-arm vibrations at work is laborious and challenging as highly specific and expensive equipment is needed. Occupational safety and health practitioners in general lack knowledge on how to perform these measurements (OSHA, 2008). Therefore, instead of objectively measuring hand-arm vibrations, assessments are often based on self-reports, guides, standardized technical reports and information provided by manufacturers. However, these sources can contain substantial errors. One reason might be that the actual exposure highly depends on the circumstances in which a task is executed, the tools that are used, the material that is processed and individual worker’s characteristics. Another reason might be that such vibration assessments are often expressed in crude, qualitative metrics. Although validity of workers’ self-reports of vibration of handheld powered tools were shown to be good to excellent (Stock et al., 2005), these estimates often systematically overestimate the actual vibration (Akesson et al., 2001; Palmer et al., 2000). Moreover, despite it has been shown that exposure to vibration should be expressed as a multiplication of the duration and the intensity of the vibration, only duration is addressed in the abovementioned studies. Therefore, reliable and valid assessment methods measuring vibration exposure in an easily applicable way at the workplace are scarce. The hand-arm risk assessment method (HARM) was developed (Douwes and de Kraker, 2009, 2014). In this assessment method, which was developed for occupational safety and health practitioners, jobs are classified according to their risk of arm, neck and or shoulder symptoms. Among other factors, such as awkward postures and duration and frequency of force exertions, HARM takes the exposure to hand-arm vibrations into account. The HARM assessment method as a whole was tested elaborately and its predictive validity has been proven for arm, neck and shoulder pain (Douwes et al., 2014; Douwes and de Kraker, 2014). However, the quality of the subjective assessment of hand-arm vibrations which is part of the HARM assessment is largely unknown. During this particular subjective assessment of hand-arm vibrations, observers classify the intensity of the vibration into one of four vibration categories (Table 1), based on the European directives on the minimum health and safety requirements regarding vibration (2002/44/EC). Therefore a simple alternative was developed that is potentially more applicable than complicated and expensive objective measurements and more accurate than self-reports or the often used data provided by the manufacturers. The aim of the present study was to evaluate the inter-observer reliability and the concurrent validity of this subjectively assessed hand-arm vibrations (as used in HARM). In this study we hypothesize that our
Table 1 HARM vibration categories and corresponding description. Category
Vibration intensity
1.
0.60 were considered good, coefficients in the range of 0.40e0.60 were considered to agree moderately and values 0.60). However, although higher agreement than the vibration intensities of the powered tools provided by the manufacturers was shown, the validity of our method was moderate, leaving room for improvement of the assessment tool. Subjective assessments can be executed in a relatively short time period (i.e., less than a minute per task) with minor recourses and background knowledge. Therefore, it can be concluded that observers without knowledge on hand-arm vibrations are able to perform an acceptable assessment of hand-arm vibrations in work situations in general. The current method can potentially be of use in risk assessments of hand-arm vibrations during industrial tasks. As can be obtained from earlier studies, potential risk of musculoskeletal complaints due to hand-arm vibrations is affected by the intensity as well as the duration of the vibration (Bovenzi, 2012; Griffin, 2004; Sauni et al., 2009). Subjective assessment of duration of hand-arm vibrations, both by self-reports and workplace observations, have been shown to be reasonable accurate before (McCallig et al., 2010; Stock et al., 2005). We add information on the validity and interobserver reliability of subjective hand-arm vibration intensity to this knowledge. The observers that assessed hand-arm vibrations had no specific experience in performing these kinds of assessments. We want our method to be easily applicable without extensive training. It can be concluded that our method is fairly applicable by observers with minor knowledge or training on the matter. However, improvement in results over time indicates a possible learning effect for the subjective assessment that apparently can be reached within a small period of time. Training of the observers may therefore improve the validity of the subjective hand-arm vibrations assessment method. Whether results found in our study would differ from data obtained in a group of trained or experienced observers cannot be concluded from our data. However, it is likely that the accuracy of subjective assessments would improve in such situations.
1261
As can be obtained from Fig. 2 and Table 4, observers had mainly difficulties to assess tasks in the second and third hand-arm vibration categories. Apparently, vibrations within these categories are difficult to distinguish. Whether these difficulties are caused by the nature of the tasks or by the description of the HARM categories cannot be concluded from the present results. A possible explanation for the abovementioned phenomenon may be a ‘floor and ceiling’ effect. As the relatively low and high vibration tasks are more likely to be assessed correctly since first and fourth category respectively, no under or overestimation is possible. However, as only two of the tasks (i.e., task 5 and 13, Table 3) show substantial high and low vibrations respectively, this effect does not seem to be of high influence on our results. Another explanation may be that the subjective assessment is difficult to perform when handling certain type of tools. Although this theory cannot be supported when comparing objectively and subjectively vibration categories obtained from the handling of tools in our study (Table 1), one cannot exclude that the handling of different tools would lead to different outcomes. A last possible explanation for the inaccuracy may be the description (e.g., hardly any vibration sensible or vibrations not visible) of the categories. Although the vibration categories are based on directives (2002/44/EC), categories might benefit from more differentiation in the description. Future research should be directed to exploring the possibilities to improve subjectively assessed hand-arm vibrations. As Fig. 2 shows that there is quite some overlap in the distributions of category 2 and 3 for objectively and subjectively assessed vibrations, revisions should mainly be targeted to improve the differentiating capacity in these categories. Improvements on the description of the assessment categories, studying different tasks and tools or training of observers might be helpful for future refinement of the method. Despite the good validity of the subjective hand-arm vibration assessment expressed in a high ICC, the percentage of exact agreement of the subjective assessment compared to the objective measurement was relatively low (i.e., 52% of all tasks). Although this agreement of 52% is higher than the 25% agreement based on chance, it should be noted that the presented assessment method is prone to improvements. Disagreement of the objective and subjective assessment of more than one category occurred only in 2 (4%) of all tasks (Table 4). It can thus be concluded that, despite a relatively low percentage of absolute agreement in absolute values, subjective assessments differ not too much from objective measurements. Furthermore, it was shown that, in tasks that were incorrectly subjectively assessed, objective measures differed 1.92 m/s2 from the category boundaries. Therefore, misclassification takes place, even when tasks are not close to category boundaries, especially in the categories 2 and 3. As can be seen in Table 4, not all categories of vibration are equally represented in our experimental protocol. Vibration category 2 and 3 occurred in 19 tasks whereas category 1 and 4 occurred in 10 and 16, respectively. This occurred as the objective vibration value per task established a-priori was not always the same as the intensity that was actually measured during the trial. Also, the objective measurement was not always assessed in the same category when repeated by the same worker. This is probably a result of the fact that the exposure to vibration highly depends on the circumstances in which a task is executed and on worker’s individual characteristics (OSHA, 2008). However, as reasonable amounts of tasks are represented in each category, it can be concluded that the present assessment method is valid and reliable in low as well as high intensity hand-arm vibration tasks. To the best of our knowledge, for ICC calculations, no particular power calculations can be performed. Therefore, we chose 64 observed tasks to be a reasonable number for this particular study design. As all calculated ICCs show to be significantly different from 0 showing
1262
P. Coenen et al. / Applied Ergonomics 45 (2014) 1257e1262
rather small confidence intervals, there does not seem to be a power issue in our experiment. A limited number of tasks and workers were assessed in our study; therefore, the external validity of the current results can be questioned. Besides, despite that we build a realistic mock-up of work situation, results might differ from results obtained in real occupational settings. However, the range of vibration during the tasks executed in our study are comparable to those measured in a large battery of industrial tasks in an earlier study (McCallig et al., 2010). Furthermore, no large differences in results can be expected when tasks would have been performed by more than two workers. From the above it can be concluded that although minor sources of biases cannot be ruled out, the present results seem applicable on a large variety of industrial tasks, workers and occupational. 5. Conclusion Our study shows that our method of subjectively assessed intensity of work-related hand-arm vibrations is fairly reliable among observers and moderately valid. The assessment method is even more valid than information provided by manufacturers that are often used in risk assessment is more easily applicable and cheaper than the more valid objective measurements. Our method for occupational safety and health practitioners, is a first attempt to use subjective risk assessments of hand-arm vibration. Regarding our results, the method can potentially be used in ergonomic risk assessments. In general, observers with minor knowledge and training on the matter are able to perform an acceptable assessment of hand-arm vibrations in work situations. Although our results are promising, results have to be reproduced under different circumstances (e.g., using different handling tools, observed by trained observers) to prove its consistency and to potentially refine subjective vibration assessments. Besides, the current assessment method is prone to some improvements that can potentially be achieved by improving descriptions of the vibration categories or training of observers, mainly in the classification of the middle vibration categories. Acknowledgments This study was financially supported by the Dutch Ministry of Social Affairs and Employment. References 2002/44/EC, 2002. DirectPatentthe European Parliament and of the Council of 25 June 2002 on the minimum health and safety requirements regarding the exposure of workers to the risks arising from physical agents (vibration). 5349 e 1, I., 2001. Mechanical vibration e Measurement and evaluation of human exposure to hand-transmitted vibration e Part 1: General requirements.
5349 e 2, I., 2008. Mechanical vibration e measurement and evaluation of human exposure to hand-transmitted vibration e Part 2-Practical guidance for measurement at the workplace. Akesson, I., Balogh, I., Skerfving, S., 2001. Self-reported and measured time of vibration exposure at ultrasonic scaling in dental hygienists. Appl. Ergon. 32, 47e 51. Bovenzi, M., 2012. Epidemiological evidence for new frequency weightings of handtransmitted vibration. Ind. Health 50, 377e387. da Costa, B.R., Vieira, E.R., 2010. Risk factors for work-related musculoskeletal disorders: a systematic review of recent longitudinal studies. Am. J. Ind. Med. 53, 285e323. Dong, R.G., Wu, J.Z., Welcome, D.E., McDowell, T.W., 2005. Estimation of vibration power absorption density in human fingers. J. Biomech. Eng. 127, 849e856. Douwes, M., de Kraker, H., 2009. Hand Arm Risk assessment Method (HARM) e a new practical tool. In: 17th World Congress on Ergonomics. International Ergonomics Association, Beijing. Douwes, M., de Kraker, H., 2014. Development of a non-expert risk assessment method for hand-arm related tasks (HARM). Int. J. Ind. Ergon. 44, 316e322. Douwes, M., Boocock, M., Coenen, P., van den Heuvel, S., Bosch, T., 2014. Predictive validity of the hand arm risk assessment method (HARM). Int. J. Ind. Ergon. 44, 328e334. Griffin, M.J., 2004. Minimum health and safety requirements for workers exposed to hand-transmitted vibration and whole-body vibration in the European Union; a review. Occup. Environ. Med. 61, 387e397. Griffin, M.J., Bovenzi, M., 2002. The diagnosis of disorders caused by handtransmitted vibration: Southampton Workshop 2000. Int. Arch. Occup. Environ. Health 75, 1e5. Hagberg, M., 2002. Clinical assessment of musculoskeletal disorders in workers exposed to hand-arm vibration. Int. Arch. Occup. Environ. Health 75, 97e105. Kittusamy, N.K., Buchholz, B., 2004. Whole-body vibration and postural stress among operators of construction equipment: a literature review. J. Saf. Res. 35, 255e261. McCallig, M., Paddan, G., Van Lente, E., Moore, K., Coggins, M., 2010. Evaluating worker vibration exposures using self-reported and direct observation estimates of exposure duration. Appl. Ergon. 42, 37e45. OSHA, 2008. Workplace exposure to vibration in Europe e an expert review. In: Administration, O.S.a.H. (Ed.), European Risk Observatory Report. European Agency for Safety and Health at Work, Luxembourg. Palmer, K., Crane, G., Inskip, H., 1998. Symptoms of hand-arm vibration syndrome in gas distribution operatives. Occup. Environ. Med. 55, 716e721. Palmer, K.T., Haward, B., Griffin, M.J., Bendall, H., Coggon, D., 2000. Validity of self reported occupational exposures to hand transmitted and whole body vibration. Occup. Environ. Med. 57, 237e241. Palmer, K.T., Harris, E.C., Coggon, D., 2007. Compensating occupationally related tenosynovitis and epicondylitis: a literature review. Occup. Med. 57, 67e74. Punnett, L., 2004. Work related neck pain: how important is it, and how should we understand its causes? Occup. Environ. Med. 61, 954e955. Sauni, R., Paakkonen, R., Virtema, P., Toppila, E., Uitti, J., 2009. Dose-response relationship between exposure to hand-arm vibration and health effects among metalworkers. Ann. Occup. Hyg. 53, 55e62. Schweigert, M., 2002. The relationship between hand-arm vibration and lower extremity clinical manifestations: a review of the literature. Int. Arch. Occup. Environ. Health 75, 179e185. Shiri, R., Viikari-Juntura, E., 2011. Lateral and medial epicondylitis: role of occupational factors. Best. Pract. Res. Clin. Rheumatol. 25, 43e57. Stock, S.R., Fernandes, R., Delisle, A., Vezina, N., 2005. Reproducibility and validity of workers’ self-reports of physical work demands. Scand. J. Work Environ. Health 31, 409e437. van der Windt, D.A., Thomas, E., Pope, D.P., de Winter, A.F., Macfarlane, G.J., Bouter, L.M., Silman, A.J., 2000. Occupational risk factors for shoulder pain: a systematic review. Occup. Environ. Med. 57, 433e442. van Rijn, R.M., Huisstede, B.M., Koes, B.W., Burdorf, A., 2010. Associations between work-related factors and specific disorders of the shoulder e a systematic review of the literature. Scand. J. Work Environ. Health 36, 189e201.
Contents lists available at ScienceDirect
Applied Ergonomics journal homepage: www.elsevier.com/locate/apergo
Validity and inter-observer reliability of subjective hand-arm vibration assessments Pieter Coenen a, Margriet Formanoy b, *, Marjolein Douwes b, Tim Bosch b, Heleen de Kraker b a Research Institute MOVE, Faculty of Human Movement Sciences, VU University Amsterdam, Van der Boechorststraat 9, 1081BT Amsterdam, The Netherlands b TNO Sustainable Productivity and Employability, Polarisavenue 151, 2132 JJ Hoofddorp, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 March 2013 Accepted 15 March 2014
Exposure to mechanical vibrations at work (e.g., due to handling powered tools) is a potential occupational risk as it may cause upper extremity complaints. However, reliable and valid assessment methods for vibration exposure at work are lacking. Measuring hand-arm vibration objectively is often difficult and expensive, while often used information provided by manufacturers lacks detail. Therefore, a subjective hand-arm vibration assessment method was tested on validity and inter-observer reliability. In an experimental protocol, sixteen tasks handling powered tools were executed by two workers. Hand-arm vibration was assessed subjectively by 16 observers according to the proposed subjective assessment method. As a gold standard reference, hand-arm vibration was measured objectively using a vibration measurement device. Weighted k’s were calculated to assess validity, intra-class-correlation coefficients (ICCs) were calculated to assess inter-observer reliability. Inter-observer reliability of the subjective assessments depicting the agreement among observers can be expressed by an ICC of 0.708 (0.511e0.873). The validity of the subjective assessments as compared to the gold-standard reference can be expressed by a weighted k of 0.535 (0.285e0.785). Besides, the percentage of exact agreement of the subjective assessment compared to the objective measurement was relatively low (i.e., 52% of all tasks). This study shows that subjectively assessed hand-arm vibrations are fairly reliable among observers and moderately valid. This assessment method is a first attempt to use subjective risk assessments of hand-arm vibration. Although, this assessment method can benefit from some future improvement, it can be of use in future studies and in field-based ergonomic assessments. Ó 2014 Elsevier Ltd and The Ergonomics Society. All rights reserved.
Keywords: Hand-arm vibration Risk assessment Validity
1. Introduction Exposure to mechanical vibrations at the workplace, such as hand-transmitted vibrations, can arise in numerous labors (e.g., in construction or manufacturing industries), for example when manually handling powered tools. These hand-transmitted vibrations are associated with a variety of signs and symptoms including vascular and neurological disorders (Griffin and Bovenzi, 2002). More specifically, it has been shown in several reviews that handtransmitted vibrations are associated with upper extremity
Abbreviations: HARM, hand-arm risk assessment method; ICC, Intra-class correlation coefficient. * Corresponding author. Tel.: þ31 88 866 52 90. E-mail addresses: [email protected] (P. Coenen), [email protected] (M. Formanoy), [email protected] (M. Douwes), [email protected] (T. Bosch), [email protected] (H. de Kraker). http://dx.doi.org/10.1016/j.apergo.2014.03.003 0003-6870/Ó 2014 Elsevier Ltd and The Ergonomics Society. All rights reserved.
complaints (Hagberg, 2002; Kittusamy and Buchholz, 2004; Punnett, 2004); for example, shoulder pain (van der Windt et al., 2000) and specific pathologies like tenosynovitis and epicondylitis (Palmer et al., 2007; Shiri and Viikari-Juntura, 2011). Although evidence is slightly inconsistent as there are also studies reporting weak evidence for the association of hand-arm vibration and upper-extremity complaints (da Costa and Vieira, 2010; Schweigert, 2002; van Rijn et al., 2010), it is generally accepted that hand-arm vibrations are an occupational risk. As a result of these potential occupational hazards, in 2002, European directives were communicated providing workers’ exposure limits for whole-body and hand-arm vibrations (2002/44/ EC). These directives, that are based on health and safety requirements, specify the maximum intensity of vibrations a worker can be exposed to, considering the duration of this specific vibration. Assessment of hand-arm vibrations is therefore based on both the duration and the intensity of the exposure. This approach is
1258
P. Coenen et al. / Applied Ergonomics 45 (2014) 1257e1262
supported by several studies showing that vibration, in which the exposure is a multiplication of duration and intensity of the handarm vibration, is associated with upper extremity disorders (Bovenzi, 2012; Griffin, 2004; Sauni et al., 2009). Therefore, duration as well as intensity of vibration should be considered to quantify the potential risk of musculoskeletal disorders of handarm vibrations. Objectively measuring hand-arm vibrations at work is laborious and challenging as highly specific and expensive equipment is needed. Occupational safety and health practitioners in general lack knowledge on how to perform these measurements (OSHA, 2008). Therefore, instead of objectively measuring hand-arm vibrations, assessments are often based on self-reports, guides, standardized technical reports and information provided by manufacturers. However, these sources can contain substantial errors. One reason might be that the actual exposure highly depends on the circumstances in which a task is executed, the tools that are used, the material that is processed and individual worker’s characteristics. Another reason might be that such vibration assessments are often expressed in crude, qualitative metrics. Although validity of workers’ self-reports of vibration of handheld powered tools were shown to be good to excellent (Stock et al., 2005), these estimates often systematically overestimate the actual vibration (Akesson et al., 2001; Palmer et al., 2000). Moreover, despite it has been shown that exposure to vibration should be expressed as a multiplication of the duration and the intensity of the vibration, only duration is addressed in the abovementioned studies. Therefore, reliable and valid assessment methods measuring vibration exposure in an easily applicable way at the workplace are scarce. The hand-arm risk assessment method (HARM) was developed (Douwes and de Kraker, 2009, 2014). In this assessment method, which was developed for occupational safety and health practitioners, jobs are classified according to their risk of arm, neck and or shoulder symptoms. Among other factors, such as awkward postures and duration and frequency of force exertions, HARM takes the exposure to hand-arm vibrations into account. The HARM assessment method as a whole was tested elaborately and its predictive validity has been proven for arm, neck and shoulder pain (Douwes et al., 2014; Douwes and de Kraker, 2014). However, the quality of the subjective assessment of hand-arm vibrations which is part of the HARM assessment is largely unknown. During this particular subjective assessment of hand-arm vibrations, observers classify the intensity of the vibration into one of four vibration categories (Table 1), based on the European directives on the minimum health and safety requirements regarding vibration (2002/44/EC). Therefore a simple alternative was developed that is potentially more applicable than complicated and expensive objective measurements and more accurate than self-reports or the often used data provided by the manufacturers. The aim of the present study was to evaluate the inter-observer reliability and the concurrent validity of this subjectively assessed hand-arm vibrations (as used in HARM). In this study we hypothesize that our
Table 1 HARM vibration categories and corresponding description. Category
Vibration intensity
1.
0.60 were considered good, coefficients in the range of 0.40e0.60 were considered to agree moderately and values 0.60). However, although higher agreement than the vibration intensities of the powered tools provided by the manufacturers was shown, the validity of our method was moderate, leaving room for improvement of the assessment tool. Subjective assessments can be executed in a relatively short time period (i.e., less than a minute per task) with minor recourses and background knowledge. Therefore, it can be concluded that observers without knowledge on hand-arm vibrations are able to perform an acceptable assessment of hand-arm vibrations in work situations in general. The current method can potentially be of use in risk assessments of hand-arm vibrations during industrial tasks. As can be obtained from earlier studies, potential risk of musculoskeletal complaints due to hand-arm vibrations is affected by the intensity as well as the duration of the vibration (Bovenzi, 2012; Griffin, 2004; Sauni et al., 2009). Subjective assessment of duration of hand-arm vibrations, both by self-reports and workplace observations, have been shown to be reasonable accurate before (McCallig et al., 2010; Stock et al., 2005). We add information on the validity and interobserver reliability of subjective hand-arm vibration intensity to this knowledge. The observers that assessed hand-arm vibrations had no specific experience in performing these kinds of assessments. We want our method to be easily applicable without extensive training. It can be concluded that our method is fairly applicable by observers with minor knowledge or training on the matter. However, improvement in results over time indicates a possible learning effect for the subjective assessment that apparently can be reached within a small period of time. Training of the observers may therefore improve the validity of the subjective hand-arm vibrations assessment method. Whether results found in our study would differ from data obtained in a group of trained or experienced observers cannot be concluded from our data. However, it is likely that the accuracy of subjective assessments would improve in such situations.
1261
As can be obtained from Fig. 2 and Table 4, observers had mainly difficulties to assess tasks in the second and third hand-arm vibration categories. Apparently, vibrations within these categories are difficult to distinguish. Whether these difficulties are caused by the nature of the tasks or by the description of the HARM categories cannot be concluded from the present results. A possible explanation for the abovementioned phenomenon may be a ‘floor and ceiling’ effect. As the relatively low and high vibration tasks are more likely to be assessed correctly since first and fourth category respectively, no under or overestimation is possible. However, as only two of the tasks (i.e., task 5 and 13, Table 3) show substantial high and low vibrations respectively, this effect does not seem to be of high influence on our results. Another explanation may be that the subjective assessment is difficult to perform when handling certain type of tools. Although this theory cannot be supported when comparing objectively and subjectively vibration categories obtained from the handling of tools in our study (Table 1), one cannot exclude that the handling of different tools would lead to different outcomes. A last possible explanation for the inaccuracy may be the description (e.g., hardly any vibration sensible or vibrations not visible) of the categories. Although the vibration categories are based on directives (2002/44/EC), categories might benefit from more differentiation in the description. Future research should be directed to exploring the possibilities to improve subjectively assessed hand-arm vibrations. As Fig. 2 shows that there is quite some overlap in the distributions of category 2 and 3 for objectively and subjectively assessed vibrations, revisions should mainly be targeted to improve the differentiating capacity in these categories. Improvements on the description of the assessment categories, studying different tasks and tools or training of observers might be helpful for future refinement of the method. Despite the good validity of the subjective hand-arm vibration assessment expressed in a high ICC, the percentage of exact agreement of the subjective assessment compared to the objective measurement was relatively low (i.e., 52% of all tasks). Although this agreement of 52% is higher than the 25% agreement based on chance, it should be noted that the presented assessment method is prone to improvements. Disagreement of the objective and subjective assessment of more than one category occurred only in 2 (4%) of all tasks (Table 4). It can thus be concluded that, despite a relatively low percentage of absolute agreement in absolute values, subjective assessments differ not too much from objective measurements. Furthermore, it was shown that, in tasks that were incorrectly subjectively assessed, objective measures differed 1.92 m/s2 from the category boundaries. Therefore, misclassification takes place, even when tasks are not close to category boundaries, especially in the categories 2 and 3. As can be seen in Table 4, not all categories of vibration are equally represented in our experimental protocol. Vibration category 2 and 3 occurred in 19 tasks whereas category 1 and 4 occurred in 10 and 16, respectively. This occurred as the objective vibration value per task established a-priori was not always the same as the intensity that was actually measured during the trial. Also, the objective measurement was not always assessed in the same category when repeated by the same worker. This is probably a result of the fact that the exposure to vibration highly depends on the circumstances in which a task is executed and on worker’s individual characteristics (OSHA, 2008). However, as reasonable amounts of tasks are represented in each category, it can be concluded that the present assessment method is valid and reliable in low as well as high intensity hand-arm vibration tasks. To the best of our knowledge, for ICC calculations, no particular power calculations can be performed. Therefore, we chose 64 observed tasks to be a reasonable number for this particular study design. As all calculated ICCs show to be significantly different from 0 showing
1262
P. Coenen et al. / Applied Ergonomics 45 (2014) 1257e1262
rather small confidence intervals, there does not seem to be a power issue in our experiment. A limited number of tasks and workers were assessed in our study; therefore, the external validity of the current results can be questioned. Besides, despite that we build a realistic mock-up of work situation, results might differ from results obtained in real occupational settings. However, the range of vibration during the tasks executed in our study are comparable to those measured in a large battery of industrial tasks in an earlier study (McCallig et al., 2010). Furthermore, no large differences in results can be expected when tasks would have been performed by more than two workers. From the above it can be concluded that although minor sources of biases cannot be ruled out, the present results seem applicable on a large variety of industrial tasks, workers and occupational. 5. Conclusion Our study shows that our method of subjectively assessed intensity of work-related hand-arm vibrations is fairly reliable among observers and moderately valid. The assessment method is even more valid than information provided by manufacturers that are often used in risk assessment is more easily applicable and cheaper than the more valid objective measurements. Our method for occupational safety and health practitioners, is a first attempt to use subjective risk assessments of hand-arm vibration. Regarding our results, the method can potentially be used in ergonomic risk assessments. In general, observers with minor knowledge and training on the matter are able to perform an acceptable assessment of hand-arm vibrations in work situations. Although our results are promising, results have to be reproduced under different circumstances (e.g., using different handling tools, observed by trained observers) to prove its consistency and to potentially refine subjective vibration assessments. Besides, the current assessment method is prone to some improvements that can potentially be achieved by improving descriptions of the vibration categories or training of observers, mainly in the classification of the middle vibration categories. Acknowledgments This study was financially supported by the Dutch Ministry of Social Affairs and Employment. References 2002/44/EC, 2002. DirectPatentthe European Parliament and of the Council of 25 June 2002 on the minimum health and safety requirements regarding the exposure of workers to the risks arising from physical agents (vibration). 5349 e 1, I., 2001. Mechanical vibration e Measurement and evaluation of human exposure to hand-transmitted vibration e Part 1: General requirements.
5349 e 2, I., 2008. Mechanical vibration e measurement and evaluation of human exposure to hand-transmitted vibration e Part 2-Practical guidance for measurement at the workplace. Akesson, I., Balogh, I., Skerfving, S., 2001. Self-reported and measured time of vibration exposure at ultrasonic scaling in dental hygienists. Appl. Ergon. 32, 47e 51. Bovenzi, M., 2012. Epidemiological evidence for new frequency weightings of handtransmitted vibration. Ind. Health 50, 377e387. da Costa, B.R., Vieira, E.R., 2010. Risk factors for work-related musculoskeletal disorders: a systematic review of recent longitudinal studies. Am. J. Ind. Med. 53, 285e323. Dong, R.G., Wu, J.Z., Welcome, D.E., McDowell, T.W., 2005. Estimation of vibration power absorption density in human fingers. J. Biomech. Eng. 127, 849e856. Douwes, M., de Kraker, H., 2009. Hand Arm Risk assessment Method (HARM) e a new practical tool. In: 17th World Congress on Ergonomics. International Ergonomics Association, Beijing. Douwes, M., de Kraker, H., 2014. Development of a non-expert risk assessment method for hand-arm related tasks (HARM). Int. J. Ind. Ergon. 44, 316e322. Douwes, M., Boocock, M., Coenen, P., van den Heuvel, S., Bosch, T., 2014. Predictive validity of the hand arm risk assessment method (HARM). Int. J. Ind. Ergon. 44, 328e334. Griffin, M.J., 2004. Minimum health and safety requirements for workers exposed to hand-transmitted vibration and whole-body vibration in the European Union; a review. Occup. Environ. Med. 61, 387e397. Griffin, M.J., Bovenzi, M., 2002. The diagnosis of disorders caused by handtransmitted vibration: Southampton Workshop 2000. Int. Arch. Occup. Environ. Health 75, 1e5. Hagberg, M., 2002. Clinical assessment of musculoskeletal disorders in workers exposed to hand-arm vibration. Int. Arch. Occup. Environ. Health 75, 97e105. Kittusamy, N.K., Buchholz, B., 2004. Whole-body vibration and postural stress among operators of construction equipment: a literature review. J. Saf. Res. 35, 255e261. McCallig, M., Paddan, G., Van Lente, E., Moore, K., Coggins, M., 2010. Evaluating worker vibration exposures using self-reported and direct observation estimates of exposure duration. Appl. Ergon. 42, 37e45. OSHA, 2008. Workplace exposure to vibration in Europe e an expert review. In: Administration, O.S.a.H. (Ed.), European Risk Observatory Report. European Agency for Safety and Health at Work, Luxembourg. Palmer, K., Crane, G., Inskip, H., 1998. Symptoms of hand-arm vibration syndrome in gas distribution operatives. Occup. Environ. Med. 55, 716e721. Palmer, K.T., Haward, B., Griffin, M.J., Bendall, H., Coggon, D., 2000. Validity of self reported occupational exposures to hand transmitted and whole body vibration. Occup. Environ. Med. 57, 237e241. Palmer, K.T., Harris, E.C., Coggon, D., 2007. Compensating occupationally related tenosynovitis and epicondylitis: a literature review. Occup. Med. 57, 67e74. Punnett, L., 2004. Work related neck pain: how important is it, and how should we understand its causes? Occup. Environ. Med. 61, 954e955. Sauni, R., Paakkonen, R., Virtema, P., Toppila, E., Uitti, J., 2009. Dose-response relationship between exposure to hand-arm vibration and health effects among metalworkers. Ann. Occup. Hyg. 53, 55e62. Schweigert, M., 2002. The relationship between hand-arm vibration and lower extremity clinical manifestations: a review of the literature. Int. Arch. Occup. Environ. Health 75, 179e185. Shiri, R., Viikari-Juntura, E., 2011. Lateral and medial epicondylitis: role of occupational factors. Best. Pract. Res. Clin. Rheumatol. 25, 43e57. Stock, S.R., Fernandes, R., Delisle, A., Vezina, N., 2005. Reproducibility and validity of workers’ self-reports of physical work demands. Scand. J. Work Environ. Health 31, 409e437. van der Windt, D.A., Thomas, E., Pope, D.P., de Winter, A.F., Macfarlane, G.J., Bouter, L.M., Silman, A.J., 2000. Occupational risk factors for shoulder pain: a systematic review. Occup. Environ. Med. 57, 433e442. van Rijn, R.M., Huisstede, B.M., Koes, B.W., Burdorf, A., 2010. Associations between work-related factors and specific disorders of the shoulder e a systematic review of the literature. Scand. J. Work Environ. Health 36, 189e201.
