Egress Efficacy of Persons with Multiple Sclerosis During Simulated Evacuations
Expedited evacuation of commercial and residential structures in the event of an emergency may be more difficult for persons with physical movement disorders. There is a need to better characterize the impact of such disorders and provide movement data to improve evacuee and responder safety. We undertook a pilot, feasibility study that investigated the ability of persons with multiple sclerosis (MS) and controls without MS to walk along a 48 m long path that included five different door configurations with various opening hardware and closure mechanisms, both before and after a six-minute walk, simulating a long evacuation path. Persons with MS took longer to complete the evacuation circuit (102 vs. 31 s) and to pass through each door (average 4.8 vs. 1.4 s) compared to controls. During the six-minute walk, persons with MS had decreased walking speed (0.7 vs. 1.9 m/s). The MS population demonstrated more conservative gait biomechanics throughout the simulation, i.e., wider, shorter and slower steps. Timing and biomechanical differences between populations and the potential fatigue induced through an extended evacuation can be used to improve understanding of movement in populations with disabilities, and incorporate these data into estimation of flow rates during evacuation.
KeywordsEvacuation Multiple sclerosis Movement disorders Gait
There are more than 9.5 million individuals in the United States with movement disorders (nearly four percent of the adult population) with that number projected to grow to 20 million (approximately five percent of the projected adult population) by 2050 with improvements in healthcare and an aging population . Movement disorders can result from a large spectrum of origins including aging, obesity, Parkinson’s disease, stroke, cerebral palsy, spinal cord injury, and traumatic brain injury among others. Many of these movement disorders are locomotion disabilities. Individuals with multiple sclerosis (MS) represent approximately 400,000 of these cases . MS is an immune response targeted at the central nervous system leading to demyelination of the nerve cells. This may lead to degradation of the motor system and affect musculoskeletal movements. Further, MS can result in significant perceptions of fatigue, even when only a mild degree of disability is present . Substantial research exists characterizing gait in the MS population [4, 5, 6, 7, 8, 9] (e.g. those with MS walk slower with a lower cadence and exhibit wider, shorter steps than controls ) and these detriments may be particularly relevant during emergency conditions.
Boyce  emphasized that individuals with locomotion and other disabilities are increasingly present in society and active in public spaces. Increased Americans with Disabilities Act (ADA) compliance provides more access for those with movement disorders to these structures. Public spaces must be designed to accommodate the egress of these individuals based on their presence and building occupancy. At the same time, public, commercial and residential structures continue to increase in size and complexity. As such longer, more complex evacuation routes with multiple door passages may be encountered in order to protect egress pathways and compartmentalize structures. It is critical to understand how these long evacuation routes with potential to encounter multiple closed doors might impact the egress capability of those with locomotion disabilities who have increasing access.
As the prevalence of movement disorders increases in society, so does the need for research regarding the evacuation capabilities and characteristics of individuals with various movement disorders. Boyce and Shields pioneered studies regarding the characterization of building occupancies for fire safety of people with disabilities [10, 11, 12, 13, 14]. Gait speed for individuals with locomotive disabilities are consistently slower than their able-bodied counterparts, with decreasing speed as the severity of the impairment increases . While studies have continued to examine movement of individuals in large and potentially complex structures such as high-rise buildings [15, 16], limited recent research has focused on evaluation of those with movement disorders [17, 18].
Recent work by the National Institute for Standards and Technology regarding evacuation for individuals with mobility impairments has emphasized the lack of existing data that can be used to estimate required safe egress time (RSET) for comparison with the available safe egress time (ASET) in fire scenarios common in today’s structures. ASET and RSET are essential components of any egress model and their accuracy must be improved for performance-based building design . These egress models often rely on available data regarding the movement capabilities, movement independence, and occupied space of the individuals within the environment. Further characterizing these parameters can result in an increased awareness of population needs and more accurate models for improved life safety .
This pilot, feasibility study is novel in that it is the first examination of gait in persons with multiple sclerosis during simulated emergencies and directly compares these experimental measurements to a control group. It was hypothesized that persons with MS would take significantly longer to complete a simulated ‘evacuation circuit’ consisting of walking through a course that included multiple door opening obstacles and that this completion time would be positively correlated with increasing levels of disability [based on the expanded disability status scale (EDSS)]. Furthermore, it was hypothesized that the MS population would exhibit wider, shorter steps than age-matched controls  and that an age-matched control group would complete the evacuation circuit slower and exhibit wider, shorter steps while travelling slower than a young control group. Lastly, the authors anticipated that, after performing a six-minute walk (simulating an extended duration evacuation), the completion time of a subsequent evacuation circuit would be slower.
2.1 Sample Descriptions
Subject Descriptives (Mean ± SD)
Multiple sclerosis (MS)
54.3 ± 5.3
83.2 ± 28.3
170.2 ± 8.8
Age matched controls (AMC)
49.7 ± 12.0
81.6 ± 19.4
171.8 ± 11.6
Young controls (YC)
21.0 ± 1.2
70.4 ± 5.8
175.0 ± 7.8
2.2 Experimental Design and Instrumentation
The experiment consisted of four phases: (1) familiarization, (2) initial evacuation circuit (EC), (3) a Six-Minute Walk (6 MW), and (4) post-walk EC. All subjects wore comfortable clothes and shoes throughout the study. The Institutional Review Board of the University of Illinois approved the protocol.
Phase 1 provided each subject with an opportunity to become familiarized with the EC arrangement, door operation, and instrumentation. The EC was demonstrated to the subject whilst the subject was seated in a wheelchair.
Each Door Presented a Unique Combination of Opening Direction, Hardware, Latch and Closer Mechaninsm and Required Opening Force
Opening Force (N)
In Phase 3, each subject completed a 6 MW to simulate an extended walking distance. The 6 MW used the EC path except that the segment to Door 5 was excluded, all doors were open, and travel was in the opposite direction (counterclockwise). The 6 MW is a reliable measure of physical capacity and fatigue during submaximal exercise [22, 23] and has been specifically validated for individuals with MS [3, 5, 24, 25, 26]. Subjects were instructed to walk as fast and as far as possible for six minutes. A researcher followed the subject with a measuring wheel (Stanley MW50, New Briton, CT) to record total distance travelled. Gait parameters were measured during the 6 MW using a gait mat and motion capture system. Metabolic data were collected using a K4b2 portable monitoring unit (Cosmed S.r.l.; Rome, Italy). At the completion of the 6 MW, subjects stopped walking and were transported by wheelchair to the start of the EC.
For Phase 4, the subject then repeated the EC as soon as possible after completing the 6 MW to characterize the potential impact of fatigue on traversing the doorways and egress after extended duration walking.
2.3 Data Collection
Total EC completion time and individual door passage times were measured during Phase 2 and Phase 4. Door passage times were recorded as the time between subject contact with the door or handle and contact of the trailing limb on the ground beyond the door threshold. EC completion time was split into door passage time (time spent passing through doors) and walking time (time travelling between doors). The sum of door passage times was subtracted from total time to determine walking time.
Gait parameters (described below) and the Oxygen Cost of Walking (O2CW) were computed from data collected in Phase 3. O2CW was computed as the steady state oxygen consumption (VO2) divided by the walking velocity and averaged over the final three minutes of the 6 MW (as in Sandroff ).
2.3.1 Spatiotemporal Gait Parameters
Velocity was calculated by the gait mat software as the forward speed of the subject over the 7.9 m length of the gait mat (GAITRite, Platinum, CIR Systems; Sparta, NJ). Gait mat data over this open stretch of the EC allow characterization of uninterrupted gait speed, which is complimentary to the total distance travelled during the 6 MW (that also includes the impact of accelerating and decelerating while turning corners, etc.). Measurement on each lap completed also provides the ability to quantify changes that may be attributed due to fatigue during the 6 MW.
2.4 Data Analysis
Data were analyzed using repeated measures ANOVAs (SPSS 23, IBM; Armonk, New York). For all measurements, comparisons were made between Groups (MS, AMC, YC). During the EC (i.e. Phases 2 vs 4), effects of Time were evaluated (pre- and post-6 MW), as well as comparisons between Doors (configuration of 5 specific doors). Both Time and Door main effects are included as repeated measures factors. Group, Time, and Door main effects and all interactions were examined where appropriate. Two-sided Pearson correlations for EC completion time and door passage time against EDSS score were completed. For the 6 MW, changes in velocity between the first and last lap (Fatigue main effect) were also studied. Significance was set at p < 0.05.
3.1 Evacuation Circuit (EC)
3.1.1 Time to Completion
3.1.2 Door Passage
3.2 Six Minute Walk (6 MW)
3.2.1 Gait Analysis
Gait Parameters During Six Minute Walk by Group (mean ± SE)
Multiple sclerosis (MS)
Age matched controls (AMC)
Young controls (YC)
Average step width (mm)
143 ± 9*
100 ± 9
103 ± 10
Maximum step width (mm)
265 ± 16*
210 ± 15
203 ± 17
Maximum step length (mm)
721 ± 41*
887 ± 40
936 ± 45
Average velocity (m/s)
0.75 ± 0.07*
1.98 ± 0.07
1.83 ± 0.08
Speed Differences Between First and Last Laps During 6 MW (mean ± SE)
Multiple sclerosis (MS)
Age matched controls (AMC)
Young controls (YC)
Velocity during First Lap (m/s)
0.81 ± 0.40
2.05 ± 0.24
1.87 ± 0.21
Velocity during Last Lap (m/s)
0.69 ± 0.34 *
1.90 ± 0.21 *
1.81 ± 0.29 *
3.2.2 Distance Travelled
3.2.3 Metabolic Analysis
Individuals with MS required significantly longer times to traverse the evacuation circuit with various doorways, as was hypothesized. That is, those with MS took more than three times as long to complete the 48 m long EC relative to the control groups (102 versus 31 s). The longer completion time in the MS group may be attributed to increased difficulty to open and pass through each door as well as changes in gait that resulted in decreased velocity of the MS group relative to the control groups (as measured during the 6 MW). These differences in walking patterns are commonly indicative of a more conservative gait, a compensation strategy that those with MS may use to maintain balance. Previous studies agree that individuals with MS walk with fewer, shorter, wider steps with a slower overall velocity [4, 6]. The increased step width suggests that individuals with MS may occupy a larger base of support and require more space to walk naturally and comfortably during an evacuation. Slower velocity in the MS group may be explained by decreased aerobic capacity [3, 6], a finding also reflected in the increased O2CW reported in the MS population in this study.
For all three groups, there was a significant decrease in velocity between the first and last lap on the gait mat during the 6 MW (Table 4). This decrease in velocity, possibly due to induced fatigue, was hypothesized; however, there were no differences in the EC completion time (Phases 2 and 4) before and after the 6 MW. Previous studies have shown that MS populations slow over the duration of the 6 MW, while healthy control groups initially slow, but return to the initial speed or greater during the final minute of the 6 MW . Perhaps there was a response within the MS population that enabled them to match initial EC completion time despite being fatigued from the 6 MW. During the 6 MW, the MS population covered a smaller distance (slower average velocity) with slower straight-line velocities. Interestingly, the average velocity across the entire 6 MW was lower than the straight-line velocities in the MS (0.70 vs 0.75 m/s) and AMC (1.86 vs. 1.98 m/s) groups, but was not different in the YC group (1.83 vs. 1.83 m/s). While turning corners and the passage through the doorways, even with the doors fixed open, resulted in decreased velocity in the older groups, the younger subjects traversed these obstacles without apparent modification to their overall speed. Importantly, the overall slower pace in the MS population can locally reduce flow of evacuees, increasing overall egress time even for those without movement disorders.
Additionally in our study, there was variability within individuals with MS, related to the severity of disease progression. EDSS is scored based on ambulatory ability, with a higher EDSS score indicating decreased mobility. High EDSS scores have been directly related to decreased 6 MW distance travelled [3, 23], and is positively correlated with time to complete the evacuation circuit and door passage time in the current study (R = 0.752 and R = 0.722, respectively). These results suggest that evacuation in an emergency scenario would take significantly longer for individuals with more severe MS who may have decreased egress capabilities. The age-matched and young controls experienced similar EC completion times, indicating that age was not a major factor for evacuation time in this study. This result is contrary to previous research indicating decreasing 6 MW distance with increased age, though age ranges in that study were from 60 to 89 . The age-matched control population in the current study averaged about 50 years, perhaps young enough to avoid the effects observed by Steffen et al.
The results from door passage time suggest that door passage is fastest when the door does not involve a latch or a handle and that traversing a push-direction door is quicker than traversing a pull-direction door. Typically, doors in common area egress pathways are designed to open in the direction of travel for safer evacuation, and the results of this study further enforce this standard. The passage times in this study are slightly faster than those reported by Boyce , perhaps because the subjects in the present study were told to navigate the doors as they would in an emergency. Boyce  reported longer passage times and increased failure rates with doors requiring greater opening force. Interestingly, in this study, there was no statistical difference between the door passage time for Doors 1 and 3, with opening forces of 33.7 N and 5.2 N, but these were significantly shorter than Door 4 that required 50.4 N of force. This study suggests that the door passage time for the MS population is more heavily influenced by direction than opening force. Additionally, while there is a trend for increasing passage time for pull doors with increasing opening force, passage time was only significantly different for the heaviest door. This finding may suggest a potential threshold value under which the ability to open and traverse the door is not affected by required opening force. Finally, while Boyce  reported some failures to operate doors at forces as low as 21 N, our population was able to open and traverse all door conditions presented.
While this study provides valuable new feasibility, pilot data on egress in MS, the sample size was relatively small. Future researchers should recruit larger and more heterogeneous samples of MS, for further understanding our novel results. The results of this study could be further expanded through a larger motion capture volume with an increased subject base to better characterize the spectrum of gait deficiencies related to MS and other movement disorders. The study population could also be expanded to include those who require the use of a wheelchair or other assist device and to include young adults with MS.
The portion of the population with movement disorders is projected to increase in future decades. As these populations become more prevalent in society, considerations must be made for individuals with movement disorders in regard to evacuation planning. This study illustrates the increased evacuation time and door passage time of individuals with MS, possibly as a result of altered gait parameters. Further, these changes become more pronounced as disease progression increases. Individuals with MS display wider, shorter steps as well as slower velocity than age-matched or young controls. Improved understanding of the spatial aspects of gait in those with MS can help in the design of evacuation simulations and building development. In this study, individuals with MS required more physical space, particularly width, during gait, which may play an important role in evacuation scenarios, when considering crowd density or the maximum capacity of an egress pathway. Overall evacuation of a structure is depended on many factors including population density and the actions of others, but improved movement data for individuals, especially those with movement disorders, is essential when considering evacuation scenarios. These data are particularly important for buildings located in areas with high populations of aging or disabled individuals, especially as these individuals gain improved access. This study can serve as an important step towards obtaining detailed evacuation characteristics for individuals with movement disorders.
Funding support for MB was provided by the National Science Foundation Engineering Research Center for Compact and Efficient Fluid Power (0540834), with additional support from the Foundation of the Consortium of Multiple Sclerosis Centers’ Multiple Sclerosis Workforce of the Future program.
Compliance with Ethical Standards
Conflict of interest
There are no conflicts of interest regarding this work.
- 2.(2005) Multiple sclerosis information sourcebook, N.M.S. Society, Editor., Information Resource Center and Library of the National Multiple Sclerosis Society, New YorkGoogle Scholar
- 16.Kuligowski ED, Hoskins BL (2010) Occupant behavior in a high-rise office building fire. NIST Technical Note 1664. National Institute of Standards and TechnologyGoogle Scholar
- 21.Justice DO (2010) ADA standards for accessible design. In: Chapter 4: Accessible RoutesGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.