Abstract
This research identifies laboratory test methods designed to advance assessment of the effects of structural firefighter gloves on a firefighter’s ability to perform tasks with their hands. Two new hand dexterity test methods are discussed: a modified tool test for measuring glove effects on gross or whole hand motor control, and a novel cylinder lift method for evaluating glove effects on fine or fingertip hand dexterity. Data generated by testing a representative group of structural firefighter and other responder gloves are used to show that these new test methods provide less variable data and a more useful and informative assessment of the effects of glove construction on hand dexterity than that provided by standard small pin pegboard tests. Based on these comparisons, a combination of the newly developed tool and cylinder lift test methods are recommended for evaluating the effects of structural firefighter gloves on hand dexterity in standards used as the basis of certifying the performance of structural firefighter clothing.
Keywords
The adverse impact of structural firefighter gloves on hand dexterity has been singled out by firefighters as the feature of their protective gloves most in need of improvement. 1 Firefighters typically compensate for the unwanted effects of gloves on their hand dexterity by carrying multiple gloves for different tasks, or by removing their gloves entirely thereby increasing their exposure risk to hand burn injuries and other environmental hazards. The availability of compliant gloves that provide an increase in hand functionality in structural firefighting operations is needed.
This research sought to identify glove dexterity test methods, or combinations of dexterity methods, which would provide more complete information on the impact of firefighter gloves on tasks that require both gross and fine hand motor control than that provided by currently used test methods. It also sought to generate relevant performance data by testing a wide range of structural firefighter gloves, and by optimizing the dexterity testing procedures themselves to provide more reliable and discriminating comparative data specifically useful for structural firefighting applications.
Methods and materials
This research evaluated and compared four different dexterity test methods. An optimized tool-type test was used to provide a work-task-simulating evaluation of glove effects on the assembly of nuts and bolts fastened using a torque and box wrench. A novel cylinder lift test was developed to assess glove effects on fine or fingertip dexterity required to grasp and lift a minimally protruding object from a flat surface. Dexterity performance ratings from these two new methods were compared with standard smooth-pin and knurled-pin pegboard dexterity tests used to evaluate the effects of structural firefighting gloves on hand dexterity.
Modified tool glove dexterity test
The Bennett Hand-Tool Dexterity Test, used in a National Fire Protection Association (NFPA) standard for structural firefighting gloves, 2 provided the basis for the first dexterity evaluation method. The Bennett Test involves using several wrenches of various sizes and types and a screwdriver to disassemble and reassemble 12 nut, bolt, and washer combinations (four each of three different sizes) onto an upright frame. Primarily designed as a tool-use proficiency aptitude test, the Bennett Test was intended to measure bare-handed gross motor skills. The 1983 Edition of the NFPA 1971 Standard adapted this test to evaluate the dexterity performance of structural firefighter gloves by comparing subject bare-handed test time to test completion with gloved hands. 3 However, a study of hand function tests, conducted by Dodgen et al., 4 identified shortcomings with adapting the Bennett Test for use by standards for emergency responder protective clothing, citing difficulty in obtaining consistent test results due to a learning curve associated with this complicated test, and issues with subject coordination and familiarity with using gloves. The Dodgen study concluded that the Bennett Test does not represent the tools or range of tasks actually performed in emergency response operations. Other issues with this method are the length of time to complete the test, the need to standardize the tightness of the nut and bolt assembly, and the impact on subject ratings from retrieving small parts or tools dropped during testing. 4
A modified tool test was recently developed by McKenna 5 that addresses these perceived deficiencies in the Bennett Test. McKenna 5 modified the tool test method specifically for evaluating the dexterity performance of structural firefighter gloves and to reduce the variability of testing results. He accomplished this by testing a range of different structural firefighter gloves to identify the best procedure for discriminating among different gloves, including optimizing the size of the nut and bolt assembly and the spacing of the holes in the test board. McKenna’s tool test was greatly simplified by limiting the nut and bolt assemblies to a single size and by reducing the overall number of nut and bolt assemblies. In addition, the size of the nut and bolt assembly was optimized for relatively thick-fingered gloves, in sharp contrast with the Bennett Test, which was designed to evaluate bare-hand aptitude in the use of tools. The lack of subject experience with using tools was addressed by reducing the number of tools to a torque wrench and a box wrench (a single tool for each hand). The tightness of the nut and bolt assembly was standardized by use of a torque wrench that produced an audible click when the nut was tightened to a predetermined level.
These protocol simplifications reduced the time needed to complete the test from well over five minutes for any of the structural firefighter gloves to just over two minutes for even the worst performing gloves. Dexterity scores were determined by comparing test completion times between bare and gloved hands. In this way, dexterity ratings were normalized based on subject aptitude for completing the protocol. An evaluation of this method showed no significant effect on dexterity ratings associated with a test subject’s experience using tools. 6
The modified tool testing procedure can be summarized as follows: the time required for a test subject to assemble and tighten a small group of bolts, nuts, and washers onto a vertical test board is measured (Figure 1).
Modified tool dexterity test. Subject-to-subject variability within the modified pegboard test using smooth pins. Subject-to-subject variability within the modified tool test.
The subject begins the test procedure by picking up the first of four bolt and washer assemblies with one hand and inserting the bolt and washer assembly into the end hole on the top row of the vertical test board. On the other side of the board, the test facilitator holds a washer, between thumb and finger, for the test subject to take and place over the end of the bolt. The participant picks up a nut and, while holding the bolt with the other hand, screws the nut onto the end of the bolt until feeling resistance. The participant subsequently grasps and uses a box wrench (or box end of a combination wrench) to secure the bolt on one side of the vertical board while using the other hand to retrieve a torque wrench to tighten the nut onto the bolt. An audible click of the torque wrench indicates that the nut and bolt have been assembled to the predetermined, 120 inch-pounds, level of tightness. This sequence of actions is then repeated for the remaining three nut and bolt assemblies, first moving across the top of the board and finishing with the bottom test hole. 6
The time to complete the test is recorded as the time from when the subject touches the first bolt and washer assembly to the time when the torque wrench clicks on the last of the four nuts. Consistent with requirements of the modified pegboard, the modified tool test procedure is repeated until the coefficient of variation of the last three repetitions does not exceed 8%. A dexterity rating is calculated as the ratio of the time required to complete the test while wearing gloves to the time required with bare hands. A lower rating in this test indicates less glove impact on hand control and better glove dexterity performance.
Cylinder lift glove dexterity test
A novel test method was specially developed to measure the effect of gloves on fine or fingertip dexterity. The cylinder lift test evaluates the isolated effect of designs on fingertip dexterity and tactility with a penalty for gloves having excessive bulk from seams, attachments, poor grip, or other hindrances at the fingertips. This method measures the minimum height (mm) that a 2-inch diameter smooth metal cylinder must protrude above a solid, flat surface before a test subject can successfully lift the cylinder from the circular hole from which it protrudes (Figure 2).
Cylinder lift dexterity test with mechanical arrangement.
A lower platform incrementally raises the cylindrical object above the upper platform until the object can be grasped and lifted by the subject.
The weight of the cylinder was an important consideration in the design of this test method. Experimentation showed that a light cylinder could easily be lifted by the test subject through positioning and using glove seams as an aid when grasping and lifting the object. A lighter cylinder did not accomplish testing goals, since it is not realistic to assume that a firefighter would be able to rely on careful positioning of the hand in order to possess manual dexterity in an emergency situation. A heavier cylinder requiring tactile sensation for lifting, but light enough such that fingers would not become fatigued, was required. The optimum cylinder weight was determined to be 1246 g.
The test is performed with both bare hands and while wearing gloves. A dexterity rating is calculated as the ratio of cylinder height with gloves to the cylinder height with bare hands. As with the modified tool and pegboard dexterity tests, a lower rating indicates less glove impact on hand control and better glove dexterity performance.
Pegboard glove dexterity tests
The NFPA 1971 Standard for Structural Firefighter Ensembles calls for the effects of gloves on hand dexterity to be evaluated using procedures described in ASTM F2010: Standard Test Method for Evaluation of Glove Effects on Wearer Hand Dexterity Using a Modified Pegboard. 7 This dexterity test method measures the time required for a subject to place 25 small pins in a pegboard with and without gloves. Two different variations of this procedure were evaluated by our research: one used smooth pins and the other used knurled pins. A smooth finish variation corresponds with the study conducted by Dodgen et al. 4 while the knurled-pin variation is specified in the 2007 Edition of the NFPA 1971 Standard. 8
Test participants
Five test subjects (four males and one female) performed each of the four glove dexterity test procedures described above. Test participants were between 18 and 25 years of age and were selected through hand measurements corresponding with a size large glove. Glove fit is an important factor affecting dexterity performance. Proper subject sizing was assured using the hand measurement procedures detailed in the NFPA 1971 Standard. 8
Test gloves
Test gloves were selected with input from firefighters and other knowledgeable individuals in the area of firefighter personal protective equipment technologies, testing methodologies, and NFPA 1971 requirements for structural firefighter gloves. Randomly evaluated in an ambient room environment (21℃, 50% relative humidity), all test gloves were commercially available and purchased off the shelf.
Test glove materials. 9
ePTFE: expanded polytetrafluoroethylene; PU: Polyurethane.
All animals refer to leather type. All leathers have a napped texture (no top grain leathers). Leathers referenced “Same as” have the same source and specification as indicated. Leathers that are the same animal type but not indicated “Same as” may or may not have similar properties.
Digitally embossed.
Simplex knit.
Interlock knit, inner side fleeced
Speciality reflective fabric; unspecified fiber composition.
Flame resistant; fiber type unspecified.
Similar but not the same as the PU barrier in Glove B.
Unspecified fiber composition.
Proximal interphalangeal.
Test glove constructions and details. 9
2D: two dimensional; NFPA: National Fire Protection Association; 3D: three dimensional; CE: European Conformity.
Certified compliant with 2007 Edition on NFPA 1971 Standard on Protective Ensembles Structural Firefighting Ensembles. 8
As was previously described by Ross et al., 9 the compliant glove samples were chosen to represent a range of state-of-the-art gloves used by structural firefighters. They incorporated different thermal liners, moisture barriers, and outer-shell components, and they used different composite layering and design concepts. Test gloves used by emergency responders and workers were also studied. These include a single-layer, pigskin work glove (Sample Q), an insulated hazmat glove (Sample R), an extrication glove (Sample V), a lightweight, thin, knit glove with a rubber-coated palm designed for lab work (Sample W), and a two-layer, leather, law-enforcement glove (Sample X). Non-compliant glove samples (Samples Q, V, W, and X) were included because they were expected to have better functional performance than those typically associated with structural firefighter gloves. Work gloves and extrication gloves are reportedly worn by structural firefighters for improved dexterity and grip performance.
Results and discussion
A modified pegboard test has been used in the NFPA 1971 Standards to determine the dexterity performance of structural firefighter gloves. In order to pass the 2007 Edition of the NFPA 1971 Standard, structural firefighter gloves must have an average dexterity rating not exceeding 250% of bare hand control.
8
The results obtained for the gloves tested using the knurled-pin and smooth-pin pegboard tests are shown in Figures 3 and 4.
Glove dexterity measured using the modified pegboard test with knurled pins where error bars show the upper and lower limits of the 95% confidence interval based on scores from five subjects. Glove dexterity measured using the modified pegboard test with smooth pins where error bars show the upper and lower limits of the 95% confidence interval based on scores from five subjects.
These data show that all structural firefighter gloves tested using the modified knurled-peg method exceed the minimum dexterity performance requirements set by both the 2000 and 2007 Editions of the NFPA 1971 Standard. They also show little difference in dexterity performance among NFPA 1971 compliant gloves, as rated by the knurled-pin pegboard method. These findings reveal the inability of the modified knurled-peg method to differentiate among a wide range of different NFPA 1971-compliant structural gloves. They also confirm the need for dexterity tests that are more responsive to differences in the design and construction of structural firefighter gloves, and for methods that identify gloves having measureable adverse impact on bare hand control in performing job-related work tasks. The smooth-pin variation produced a considerably larger range of dexterity ratings from the test gloves but was also much more variable, leading to the same set of problems.
Comparisons using newly developed dexterity test methods
Figures 5 and 6 show test glove dexterity ratings obtained using the new cylinder lift and modified tool dexterity tests.
Glove dexterity measured using the modified tool test where error bars show the upper and lower limits of the 95% confidence interval based on scores from five subjects. Glove dexterity measured using the cylinder lift test where error bars show the upper and lower limits of the 95% confidence interval based on scores from five subjects.
Summary results from dexterity test methods (National Fire Protection Association (NFPA) 1971-compliant gloves)
All test gloves, including NFPA 1971-compliant and non-compliant responder and work gloves.
NFPA 1971-compliant structural firefighter gloves only.
Average coefficient of variation among subjects evaluating each glove.
Based on a Fisher Least Significant Difference analysis using α = 0.05.
These data show that a significant range of different dexterity ratings are obtained for the test gloves, depending on the particular test method used for the evaluation. The most agreement among the different dexterity test protocols is observed for gloves at the extreme ends of performance ratings, or gloves having the best or the worse dexterity performance. For example, all four dexterity tests are able to identify the superior performance of gloves used in applications other than structural firefighting. These include the law-enforcement glove (Sample X), the work glove (Sample Q), the extrication glove (Sample V), and the lightweight, knit glove (Sample W). Likewise, these tests revealed the poor dexterity performance of the hazmat glove (Sample R). However, differences in dexterity ratings can otherwise vary significantly by test method, especially among the group of structural firefighter gloves.
Differences observed in dexterity ratings produced by these test methods must be weighed in light of the variability of the test data. In this regard, one of the most important findings of this study is related to differences in the range and variability of dexterity ratings produced by the different test methods. A high level of testing variability was found in all the dexterity test methods studied, with overall coefficients of variation ranging from about 13% to about 24% (Table 3). Figures 7–10 show that much of the testing variability can be attributed to subject-to-subject variability in dexterity rating data. This variability was determined using the Connecting Letters Report output from the JMP® statistical software one-way analysis of variance (ANOVA). The means comparisons are made for each pair using Student’s t-test analysis with α = 0.05. Levels not connected by the same letter are statistically different.
Subject-to-subject variability within the cylinder lift test. Subject-to-subject variability within the modified pegboard test using knurled pins.
These data show that the pegboard test variant that uses smooth pins, while detecting the widest range of glove dexterity performance ratings, produces the most variable test data in comparison to the other dexterity methods evaluated by this study. This finding is consistent with our laboratory’s observation that pegboard tests characteristically involve distinct learning curves. While still adhering to instructions prescribed by the standard, a test subject learns and subtly adapts their technique for pinching and manipulating small pins, especially when wearing structural firefighter gloves with poor fingertip tactility. This problem is compounded by subject fumbling or dropping pins, and as they become frustrated and physically fatigued by repeated attempts to pick up and manipulate the small pins. We observed that the use of knurled pins alleviates, but does not eliminate, subject fumbling and frustration during pegboard tests.
Subjects found it easier to pick up knurled pins versus smooth pins. Picking up smooth pins requires some degree of fingertip tactility and grip. In comparison, picking up knurled pins requires less tactility and is not sensitive to glove surface properties. Analysis of the residuals of a fitted line between knurled and smooth pegboard results reveals that test gloves with the slickest material in the fingertips (R, S, and T) perform better in the knurled-pin test than in the smooth-pin pegboard test. Therefore, gloves S and T received a higher comparative rating for the knurled-pin compared to the smooth-pin pegboard test (see Figures 4 and 5), despite the fact that they are designed using multiple layers, tabs in fingertip construction, and a smooth textured outer shell layer in the fingertips.
Gloves S and T were generally regarded as good overall gloves for dexterity based on subjective rankings, although evaluators found that they had poor fingertip sensation. The hazmat glove (Sample R), on the other hand, was universally ranked as the worst glove for dexterity by all test subjects. While it scored poorly in the knurled pegboard test, it performed better in the knurled-pin pegboard test than in the smooth-pin pegboard, the cylinder lift, and the modified tool test (see Figures 4–10). These findings reveal test insensitivity in the knurled-pin pegboard method when evaluating fingertip dexterity of certain thicker, multi-layer gloves. The knurled-pin test tends to overrate the dexterity performance of bulky and slick finger gloves because it is easy for a knurled pin to adhere or bite into surface of the glove material used in the fingers. Consequently, the subject can pick up a knurled pin without actually feeling the pin itself. The apparent loss of this important component of tactile sensation may contribute to the observed inability of the knurled-pin pegboard test method to discriminate among different firefighter gloves based on hand dexterity function (Figure 4).
Rank correlation of subjective measurements and objective measurements
Correlation between objective tests and subjective perception of glove dexterity
After completing all dexterity tests, the test participants were asked to rank test gloves from best to worst, based on their subjective perception of overall dexterity performance. Table 4 shows the rank order correlations that were observed.
These data indicate that, when the entire glove test group is considered, the cylinder lift method is more highly correlated with perceived glove dexterity performance than either the tool or pegboard tests. The cylinder lift is primarily a test of fingertip bulk. Firefighters have attributed much of their perceived poor dexterity to the bulkiness of their gloves. 1 This finding, therefore, is an indication that fingertip bulk is either associated with overall bulk or is an aspect of bulk that limits perceived dexterity.
Although results from the cylinder lift test correlate reasonably well with the subjective perceptions of dexterity, this test cannot account for all aspects of dexterity: it does not measure the effects of the stiffness or flexibility of the glove on gross hand movements. The modified tool test is the only one of these test methods that can be used to evaluate effects on gross hand movements.
Correlations between dexterity test methods
Correlation of dexterity measurements for various test methods
These data indicate correlations between these dexterity tests when all gloves are considered. At the same time, little correlation between the methods is indicated, when considering only the NFPA 1971-compliant structural firefighter gloves. The observed lack of correlation between the tool dexterity test and pegboard and cylinder lift tests can be explained as follows: the modified tool test involves use of the whole hand to manipulate a torque wrench and fingertips to grasp and manipulate relatively smaller nut and washer assemblies. It therefore provides an evaluation of glove effects on both gross and fine hand control. In comparison, the cylinder lift and pegboard tests measure aspects of fine hand control and the effect of gloves on the ability to grasp small objects with the fingertips while wearing gloves.
Conclusions
This research has demonstrated that an optimized tool test method and a novel cylinder lift dexterity test method developed by this research provide more informative and less variable performance data for evaluating the effects of structural firefighter gloves in comparison to standard pegboard-type methods. The improvements in testing variability can be attributed to reduction in learning time and to a reduction of subject physiological frustration and hand fatigue inherent in manipulating small pins in pegboard procedures. This study has revealed the shortcomings of relying on small-pin pegboard tests to evaluate structural firefighter gloves: the smooth-pin pegboard protocol is too variable to provide reliable performance rating data for thick multilayer gloves. The knurled-pin variant of the pegboard method improves testing variability at the expense of tactile sensitivity. This testing artifact means that test subjects do not rely on tactile feedback (for thick multilayer gloves) to pick up the knurled pins. This is a serious flaw in this method that directly leads to overrating the fingertip dexterity of bulky or slick finger glove constructions. This finding helps explain the persistent firefighter complaints about the dexterity performance of their NFPA 1971-compliant gloves in field use. It shows that the performance requirement based on the knurled-pin pegboard method not only provides little basis for differentiation among structural firefighter gloves, but it may also provide limited or even misleading information about the true effects of structural firefighter gloves on fine, or fingertip, dexterity. This conclusion is strongly supported by the data generated in this study, which shows that the knurled-pin pegboard method provides little basis for differentiating among a diverse group of structural firefighter gloves. All of the NFPA 1971-compliant gloves tested easily passed the dexterity performance requirement in the 2007 Edition of The NFPA 1971 Standard based on the specified knurled-pin pegboard test performance requirement of 250% of the bare hand dexterity rating. The performance requirement for hand function in the NFPA 1971 Standard has since been raised to a 220% rating in the knurled-pin dexterity test. 11
This research has shown that more than one test is needed to provide the most complete and useful information about the effects of multilayered structural firefighter gloves on hand control. One of the tests should measure glove effects on gross or whole hand dexterity. McKenna 5 has developed a modified tool glove dexterity test that has been demonstrated by this research to effectively serve this purpose. This tool assembly set-up and protocol correct shortcomings in the Bennett Test and is optimized to differentiate the dexterity performance of a range of modern structural firefighter gloves in an evaluation that requires just one to two minutes per subject evaluation.
Finally, this research has shown that the described optimized tool test provides glove dexterity performance data that is fundamentally different from fingertip dexterity measures, such as pegboard or cylinder lift tests. Of the fingertip dexterity tests, the new cylinder lift method has emerged as a significant improvement over pegboard-type tests, primarily because of less test variability and the elimination of a serious testing artifact associated with the knurled-pin pegboard method. The cylinder lift test is not a time-based test that is subject to the learning curves or psychological effects that afflict other dexterity tests. While it cannot alone measure all aspects of dexterity, it effectively isolates the effects of glove construction on fingertip dexterity.
In summary, this research study has provided important technical data supporting the value of employing a modified tool test for evaluating the effects of structural gloves on hand dexterity. It has provided a basis for combining the use of the modified tool test with the use of a newly developed cylinder lift fingertip dexterity test. It has provided data for considering performance criteria that may be applied with these test methods by testing the widest and most representative range of glove samples that has ever been included in a study focused on the dexterity performance of structural firefighter gloves. These improvements in test methodologies and associated performance criteria will not only advance performance standards for structural firefighter gloves, but they should also contribute to the development and fielding of NFPA 1971-compliant gloves having dexterity that is more acceptable to firefighters.
Caveat
This paper describes the results of a limited laboratory study designed to provide an initial basis for understanding technical aspects of glove performance. Care must be taken in drawing conclusions about the safety benefits from these data. The data describe the properties of selected gloves in the controlled laboratory tests that are specified. Study results must be weighed in light of the fact that no laboratory analysis can completely qualify complex firefighting events, which can be physically complicated and unqualified. This study was not intended to recommend or exclude any materials from any particular application.
Footnotes
Acknowledgement
Assistance was provided by the panel of experts organized by the Fire Protection Research Foundation and the NFPA 1971 Task Group on Gloves.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Department of Homeland Security and the Federal Emergency Management Agency Assistance to Firefighters Grant Program-Fire Prevention and Safety Grants (EMW-2008-FP-00664).