Relating Real-World Falls to Laboratory Test Conditions

Introduction: : Falls cause over 50% of fatal work-related traumatic brain injuries (WRTBIs), urging improvement in head impact safety measures. Current construction industry guidelines established by the Occupational Safety and Health Administration (OSHA) mandate the use of head protection when there is risk of head impact and fall protection when working over a height of 6 ft. However, inadequate use of fall protection remains the most cited OSHA violation, indicating safety helmets may be the only protection for workers from falls. ANSI standard Z89.1-2014 specifies an impact velocity of 3.5 m/s during safety helmet standard testing, which does not adequately represent energy levels associated with falls. The understanding of real-world head impact energies associated with falls in industrial settings is limited, though necessary to best represent real-world falls in standardized laboratory tests.

Real-world head impact energies are difficult to quantify because of fall impact mitigating mechanisms (FIMMs), including active maneuvers like bracing and rolling, and passive mechanisms like sequential body impact where other body parts strike the ground first. Each of these mechanisms act to slow the head’s velocity before impact. Little is known about the quantitative reductions in head impact velocity due to FIMMs. To more comprehensively test safety helmets under realistic impact scenarios, estimates of head impact velocity from falls accounting for FIMMs are needed. This study aimed to estimate the relationship between fall height and head impact velocity during real-world falls accounting for FIMMs and use these estimates to quantify the effects of FIMMs on head impact velocity.

Materials and

Methods: : First, a nominal head impact velocity was selected (V_laboratory). Second, the peak linear acceleration (PLA) that corresponded to this head impact velocity was determined from a linear regression line created from a series of laboratory drop tests using an instrumented, 50th percentile male NOCSAE headform in bare and helmeted conditions to front, back, and side locations (Fig. 1). Third, the skull fracture risk associated with this PLA value was determined using a skull fracture risk curve generated from a previously published dataset of post-mortem human subject head impacts. Fourth, the fall height (h), measured from the feet, corresponding to this same skull fracture risk was determined using another skull fracture risk curve created from OSHA accident reports. The relationship between head impact velocity and fall height established in these initial four steps included any effects of FIMMs, as skull fracture risk from OSHA reports were assessed from actual fall events. Step five of our analysis determined the head impact velocity from the same height without the presence of FIMMs (i.e., a head-first fall) using the theoretical kinematics of a falling particle mass with no air resistance (Eq. 1). The added height of a 50th percentile male was included to account for the added potential energy to their head by standing. Lastly, the effect of FIMMs were quantified as the percent reduction in head impact velocity between the theoretical worst-case value and the laboratory estimated value from the initial four steps (Eq. 2).

Results, Conclusions, and Discussions:: The linear (bare) and exponential (helmeted) regression lines for the head injury metric data vs. the measured velocity yielded R-squared values of 0.87 and 0.85, respectively (Fig. 1). An exponential fit better suited the helmeted data, as head injury metrics increased drastically when the helmet reached its protective limit at high impact velocities. Despite this, the helmeted data averaged lower PLA values than the bare.

The PLA skull fracture risk curve demonstrated a logistic growth trend, while the fall height risk curve exhibited logarithmic growth, increasing sharply at low fall heights (Fig. 2). Even small falls, as indicated by real-world data, have the potential to be injurious when considering additional potential energy to the head due to a person’s height. There was substantial overlap in fall heights resulting and not resulting in skull fracture, indicating several other factors other than fall height contribute to a person’s skull fracture risk, such as body orientation, impact surface, head protection, and individual head injury tolerances.

The average effects from FIMMs when examining the bare and helmeted data were 64.4% (CI 63.1 – 65.7%) and 51.5% (CI 47.6-55.3%), respectively. FIMM effects in both cases increased as fall height increased. We estimate that, on average, FIMMs can reduce a person’s worst-case head impact velocity by 44.5 – 64.4%. Laboratory head impact velocities were lower than the worst-case fall velocities not accounting for FIMMs (Fig. 3). Effects from FIMMs for the bare head condition were consistently higher than the helmeted condition at matched fall heights. At lower heights, FIMM effects differed more between the two conditions, with a maximum difference of 25.0%. As fall height increased, the effects from FIMMs became more similar, differing by 10.9% at 20 ft.

This study estimates how much FIMMs can reduce a person’s free fall velocity and can guide impact energy level selection when testing industrial safety helmets. Safety helmets that are designed and tested under realistic head impact velocities and fall conditions have the potential to improve their real-world protective capabilities and decrease the number of WRTBIs.

Acknowledgements and/or References (Optional):: V_worst-case = sqrt(2*g*(h+1.75)) (Eq. 1)
FIMM Effects = (V_worst-case - V_laboratory / V_worst-case) * 100% (Eq. 2)