Slip-and-falls account for a significant proportion of workplace injuries but can be ameliorated by footwear with high friction. The rate of nonfatal occupational injuries due to falls has been consistently in the range of 25% to 27% between the years of 2011 and 2018 (U.S. Department of Labor- Bureau of Labor Statistics, 2020a, b). Slips are the initiating event for around 50% of these injuries (Courtney et al., 2001). Slip-resistant footwear has emerged as an effective intervention in preventing slips although the efficacy of this footwear is reduced as the shoe ages (Bell et al., 2019; Verma et al., 2011; Verma et al., 2014). Recent research has demonstrated that programs that take an active role in selecting high performance footwear can reduce slip events (Bell et al., 2019). However, best practices for this type of program are still emerging.

The design of the shoe tread is known to influence an individual’s slip risk. Coefficient of friction (COF) can vary dramatically across footwear designs (Beschorner et al., 2019a; Grönqvist, 1995; Grönqvist et al., 1993; Iraqi et al., 2018; Iraqi et al., 2020; Jones et al., 2017; Yamaguchi et al., 2015; Yamaguchi et al., 2017). Outsole parameters including contact area (Iraqi et al., 2020; Jones et al., 2018), tread channel features (depth and width) (Beschorner and Singh, 2012; Blanchette and Powers, 2015; Li and Chen, 2004, 2005), and material hardness (Cowap et al., 2015; Iraqi et al., 2020; Jones et al., 2018; Moore et al., 2012; Strobel et al., 2012) have been shown to influence COF. Furthermore, naturally worn (Grönqvist, 1995; Hemler et al., 2018) and artificially worn shoes (Beschorner and Singh, 2012; Gao et al., 2003; Hemler and Beschorner, 2017) have demonstrated a reduced shoe-floor COF on oily surfaces relative to their new counterparts. Lower COF values are associated with higher slip risk (Beschorner et al., 2019a; Hanson et al., 1999; Iraqi et al., 2018). Thus, identifying footwear tread features that enhance friction and thus lower slip risk is critical to preventing slip events.

Hydrodynamic theory can be used to explain the relationship between worn shoes and lower COF. In hydrodynamic lubrication, the fluid between the two surfaces becomes pressurized, which causes the surfaces to separate and the COF to decrease (Hamrock et al., 2004). Substantial under-shoe fluid pressures have been observed for worn and partially worn shoes in the presence of high viscosity contaminants (Beschorner et al., 2014; Beschorner and Singh, 2012; Hemler et al., 2019a; Singh and Beschorner, 2014). The Reynolds equation describes the relationship between fluid pressure and the separation between the surfaces (Beschorner et al., 2009; Chang et al., 2001b; Hamrock et al., 2004). These pressures can develop due to converging surfaces in motion (the wedge effect) (Beschorner et al., 2009; Beschorner et al., 2014; Chang et al., 2001b; Moore et al., 2012; Proctor and Coleman, 1988) or the squeeze-film effect (Beschorner et al., 2014; Chang et al., 2001b; Grönqvist et al., 2003; Strandberg, 1985), which are both believed to apply to shoes. Both the wedge and squeeze-film effects predict that fluid pressurization is sensitive to the size of the interacting surfaces (Hamrock et al., 2004; Hemler and Beschorner, 2017; Proctor and Coleman, 1988; Strandberg and Lanshammar, 1981). This concept has been recently applied to shoes by quantifying the size of the fully worn region. Specifically, shoes with a larger worn region were found to have higher fluid pressures and lower COF (Hemler et al., 2020a; Hemler et al., 2019a). Thus, both hydrodynamic theory and experimental studies suggest that larger worn regions have a negative influence on a shoe’s friction performance.

Few methods are available to determine when a worn shoe should be replaced. Two potential metrics have been suggested for tracking shoe wear: shoe age and shoe condition. Slip-resistant footwear companies often recommend replacing shoes older than six months and replacing shoes is associated with reduced slip risk (Verma et al., 2014). One limitation of this approach is that individuals wear through shoe tread at different rates depending on their gait style and usage (Hemler et al., 2019b; Pliner et al., 2019). A second approach is to determine replacement thresholds based on the condition of the tread. For example, one footwear company provides a tool for assessing tread depth, which has an indicator for when to replace the shoes (Shoes for Crews, 2019). This method mirrors tread gauges, which are commonly used to assess tire tread (Gunaratne et al., 2000; Tracy et al., 2004). A potential limitation to this approach is that tread wears unevenly across the shoe surface (Grönqvist, 1995; Hemler et al., 2019a; Hemler et al., 2018) and the tread depth measurement is dependent on the location. Presumably, the tool could be applied to the location of least depth. Yet localized wear can occur rapidly and reductions in shoe friction performance occur substantially after a region of the shoe becomes fully worn (Hemler et al., 2019a). Therefore, using the minimum tread depth may be overly conservative and tread gauges may not translate very well to shoes. Given recent evidence that the size of the heel worn region is predictive of friction performance (Beschorner et al., 2019b; Hemler et al., 2019a), tools that assess the size of the worn region may have potential in determining if a shoe should be replaced.

Inexpensive and easy-to-use ergonomic tools tend to be frequently used among ergonomists or safety practitioners. The choice of ergonomic tool mainly depends on the nature of an injury risk, practicality (e.g., time, cost, availability), and an ergonomist’s preference (e.g., expertise, experience) (Dempsey et al., 2005). A survey of ergonomics practitioners revealed high usage rates (60–87%) of observational assessment methods such as the NIOSH Lifting Equation (87%). However, a much lower usage rate (15–29%) was found for instrumentation-based methods such as the trunk electrogoniometer (e.g. Lumbar Motion Monitor) (18%) (Lowe et al., 2018). The higher usage of observational-tools over instrumentation-based methods is likely because observational tools cost less and require less expertise to use (Dempsey et al., 2005). Given the under-usage of slipmeters or tribometers (14.8%) among safety practitioners (Lowe et al., 2018), development of easy-to-use and inexpensive tools may increase adoption of shoe traction assessments. Furthermore, shoe tribometers (i.e., whole-shoe testers) are likely less utilized than floor tribometers since shoe tribometers (Chang et al., 2001a) are not typically portable and are relatively expensive (Iraqi et al., 2020). The lack of inexpensive and easy-to-use tools to assess the friction of shoes is a barrier for safety practitioners to assess footwear friction. Thus, the development of a new ergonomic method that meets the attributes of an ideal ergonomic tool (i.e., predictive, robust, quick to administer, inexpensive, non-invasive and easy-to-use) (Marras and Karwowski, 2006) could have an impact on reducing the magnitude of occupational slip and fall injuries.

The purpose of this study is to determine the ability of a simple observational test (i.e., appropriate ergonomic tool) that uses a readily available object (alkaline batteries) to assess wear-related loss of friction performance in slip-resistant shoes based on the size of the worn region. Under-shoe fluid pressures will also be analyzed to assess the role of fluid pressurization in shoe performance.

This study demonstrates that the wear-induced reduction in fluid drainage capacity and friction in slip-resistant shoes can be assessed by comparing the size of the worn region to an object such as a AAA and AA battery. Consistent trends were observed in data from the three different experiments, including: a controlled accelerated wear procedure; a longitudinal study tracking shoe performance throughout its naturally-worn life; and a cross-sectional study examining human exposures to slippery contaminants, while wearing naturally-worn shoes. Specifically, a reduction of approximately 30% in COF was observed for shoes failing these tests relative to their passing counterparts. Higher fluid pressures were consistently observed in cases where shoes failed the test, however the magnitude of this difference varied across experiments. For the human slip study, increased slipping and slip severity were observed for participants wearing shoes that failed the test compared with participants wearing shoes that passed the test. Although, the battery tests were not fully deterministic of shoe friction performance, they were highly predictive in loss of friction performance within the longitudinal experiments. In fact, every shoe that was tracked during wear (shown in Figure 2A) had less friction for the fail condition relative to the pass condition. Furthermore, the tests (particularly the AA test) were moderately sensitive and highly specific predictors of human slipping. The consistency of the results across these three experiments yields robust evidence that applying these tests are an effective way of identifying worn shoes that have degraded performance. However, potential users should be trained to properly apply these tests and understand their scope of applicability. Both tests (AAA and AA) were demonstrated to predict a degradation in friction performance. The AAA test will lead to more frequent replacement of footwear than the AA test. Therefore, the AAA test is an option for more risk-averse practitioners, while the AA test may be more appropriate to those who are more risk-tolerant.

The proposed battery tests are designed to be a simple and inexpensive solution in assessing the worn condition of slip-resistant shoes. The batteries are a consistent and available product that can lead to wide adoption to assess tread wear and prevent slip-initiated falls. These features are consistent with attributes of an ideal ergonomic tool (i.e., predictive, robust, quick to administer, inexpensive, non-invasive and easy-to-use) (Marras and Karwowski, 2006). While this test is easy to perform, future research should examine the human factors aspects of this test including inter-rater reliability, usability, and usefulness of the test.

This test assesses the actual worn condition of shoes as opposed to other cutoffs like the age of the shoe. Commonly, shoes are replaced according to a time schedule (e.g., every 6 months) (Verma et al., 2014). However, shoe age guidelines do not account for variation in wear due to differences in individual usage or biomechanics. This study found that a substantial portion of shoes with less than 6 months of use failed one of the battery tests (27–36% from Experiment #2) and a substantial portion of shoes older than 6 months passed the test (56–70%). Thus, discrepancies exist between time-based replacement thresholds and worn-condition thresholds. Presumably, replacing shoes on a set time-based schedule may expose high wear individuals (high use or gait pattern that is associated with faster wear (Hemler et al., 2019b)) to increased slip and fall risk and may lead to unnecessary cost for low wear individuals (lower use or gait pattern associated with slower wear). Thus, the utilization of this test may enable reallocation of resources to the individuals at greatest slipping risk.

The results of this study are consistent with previous research that demonstrated degraded performance for worn shoes. Our previous research has found that an increase in fluid pressures and a reduction in COF is associated with shoe wear (Beschorner et al., 2014; Beschorner and Singh, 2012; Hemler and Beschorner, 2017; Hemler et al., 2018; Singh and Beschorner, 2014). In addition, we also showed a relationship between the size of the worn area and friction (Beschorner et al., 2019b; Hemler and Beschorner, 2017), which is consistent with assessments targeting the size of the worn region. Thus, the results of these experiments are generally consistent with previously-identified trends of the influence of shoe wear on slip risk.

This study focused on wear in the heel region of the shoe as opposed to the midfoot or forefoot regions. The heel is appropriate for tracking wear since previous research has identified that wear commonly occurs in this region (Hemler et al., 2020b). Furthermore, slips occurring during the heel contact phase of gait are relevant to fall risk since they are 1) capable of generating a backward fall away from the point of contact; 2) occur during the initiation of swing phase for the contralateral limb; and 3) have high frictional requirements (Redfern et al., 2001). While wear in other regions of the shoe may lead to loss of friction, wear in the heel region is more likely to lead to slips resulting in a fall. Thus, observing wear in the heel region is justified for achieving the goal of reduced slip and fall events.

Certain limitations of this study should be acknowledged regarding the scope of applicability. First, the study focused on shoes that were labeled as “slip-resistant” by the manufacturers and it is unclear if the same test is valid across all shoe types. Second, although this test was effective at predicting loss of friction, it was not fully deterministic of friction performance as indicated by the variability in COF across shoes (Figure 2). Other parameters are known to be relevant to friction performance including tread surface area, material hardness, heel shape, and tread bending stiffness (Iraqi et al., 2020; Jones et al., 2018; Yamaguchi et al., 2017). Third, while the threshold was tested across data from multiple experiments using a variety of slip-resistant shoes, further efforts are needed to generalize the results to different contaminants and flooring. Tribology theory suggests that the worn regions assessed in this study would lead to greater hydrodynamic pressures and lower COF values when utilized with lower roughness flooring and higher viscosity contaminants (Hamrock et al., 2004). Thus, some correction factor may be warranted when applying these results to smooth floor surfaces with high viscosity contaminants. Lastly, validating these tests using actual workplace slip and fall data would increase confidence in the test.

This study demonstrates that worn shoes, as assessed using a simple test of comparing the worn region to the size of batteries, are associated with a loss in friction. This result was demonstrated consistently across three experiments and is consistent with previous research and theory. The test is expected to offer workers a simple, quick and inexpensive way to assess their shoe wear condition and enable employers to appropriately allocate resources based on the actual worn condition of the shoes. Broader use of shoe wear monitoring through the proposed tests has an opportunity to reduce the burden of slip and fall events in the workplace.