Exploring fall training adaptations while walking

Abstract

BACKGROUND: Trips are common in and out of the workplace with most people recovering to avoid a subsequent fall. However, when the recovery attempt fails, a fall can be detrimental.

OBJECTIVE: The purpose of this exploratory study was to examine adaptations to the elevating response during obstacle tripping while walking on a treadmill. Additionally, the possible transfer effects from adapted responses in the lab to the worksite are explored.

METHODS: Fourteen healthy participants that covered the general working age range (20–70 yrs.) were presented with two different types of tripping obstacles while walking.

RESULTS: Elevating the foot over the obstacle was expected due to all trips being induced during early swing phase (first 33% of the swing phase). However, in addition to the elevating strategy, a novel “push” strategy was observed in all but three participants.

CONCLUSION: The current study provided support that obstacle type influences the behavioral response after a trip. Therefore, obstacles that catch the shoe should be considered when designing functional fall programs. Furthermore, information from the current study is useful for establishing guidelines when developing a fall prevention program in the workplace.

Keywords

Obstacle training workplace stumbles push response trip response strategies

1 Background

Trips are common in and out of the workplace with most people recovering to avoid a subsequent fall. However, when the recovery system fails a fall can be detrimental. According to the U.S. Bureau of Labor Statistics, 699 workers died from a fatal fall, slip, or trip in 2013 [1]. Furthermore, the annual cost associated with non-fatal falls has been estimated to be 19 billion dollars [2]. During the construction of the Denver International Airport, falls accounted for 25 percent of workers’ compensation claims (more than 10 million dollars) [3]. Outside of the workforce, in 2003, 1.8 million older adults (65+ years) experienced a fall that resulted in treatment [4]. Twenty to 30 percent of falls in older adults caused moderate to severe injuries such as bone fractures or head trauma [5].

In addition to financial burdens linked with fall injuries, there are mental ramifications of falling, such as fear of falling [6]. Skeletal muscle atrophies with increased age [7] and is accompanied by a subsequent decrease in physical ability that can lead to a fear of falling. Fear of falling disrupts activities of daily living as well as causing an individual to change his or her natural gait pattern to one that is relatively more stable but perhaps less efficient in the work place (i.e. shorter, wider steps) [8].

Unrecovered trips account for falls. A trip causes the center of mass to be displaced toward the anterior region of the base of support. Previous research by Eng, Winter, and Patla, [9, 10] successfully classified the human recovery responses following a trip while walking. After contact with an obstacle, while walking, the swing limb either lowers in front of (lowering response) or elevates over (elevating response) the obstacle [9]. The lowering response occurs when the swing limb is perturbed during the late stages of the swing phase. The elevating response occurs when the swing limb is perturbed early in the swing phase [9]. These two responses occur following a trip while walking in the non-amputee population [11].

During the elevating response, the stance limb elevates the swing limb by plantar flexion [9]. In addition to elevation, a delayed toe off occurs prolonging stance duration by approximately 70 ms [9]. The stance ankle joint moment increases recovery time for proper positioning by decelerating the angular momentum of the center of mass [12]. Activation of the posterior muscles of the stance limb is similar for early and late swing phase perturbations [9], and these muscles stabilize the trunk when recovering from a trip. To avoid a fall, improved recovery strategies are necessary. The lowering and elevating strategies are functionally appropriate responses to a trip [9 , 14]. Improving these recovery responses may decrease falls in the workplace.

Currently, many fall prevention programs focus on improving strength, flexibility, and/or decreasing environmental hazards (i.e. selecting the right floor surface, floor maintenance, and periodic inspections) in order to increase the likelihood of recovering from a fall or to avoid the fall altogether [15, 16]. However, four fundamental problems exist with many current fall prevention approaches. First, adherence to certain health promotion programs, including physical activity, is limited [17]. Second, the intensity of many fall prevention programs is low [18] and do not provide the needed improvements. Third, some clinical and laboratory assessments and analyses cannot predict fallers from non-fallers [19]. Fourth, many current assessments and analyses do not take into account the biomechanics of posture and balance [20].

Functional fall training is a new promising method for fall prevention after a trip. This intervention can increase participant adherence by decreasing the amount of total training time required to improve balance recovery skills. Functional fall research is specific to trips by providing an intense ecologically valid perturbation in order to improve the elevating and lowering responses. Therefore, this training provides the participant with multiple controlled trips in a relatively short amount of time. Research utilizing this training has demonstrated improved recovery after trips in both older and younger age groups [21 –23]. However, minimal research examining possible adaptations that occur to the elevating response following trip training on a treadmill when obstacles are used to induce a trip exists. For example, in a pilot study we conducted, healthy college age participants learned to adapt rapidly from a trip, using a trip board, on a treadmill (see Fig. 1). In the first few trials, the expected recovery response (elevating the foot to clear a trip board during an early swing phase trip coupled with control of forward pitch of the trunk) changed. Instead, participants dropped the perturbed foot to the treadmill belt after contact with the trip board and allowed the belt to carry them backwards before attempting to move the limb again. This adaptation demonstrated that the lowering response after contact with a stationary obstacle on a treadmill was the effective recovery response. However, based on the phase of the gait cycle (i.e. early swing phase), the lowering response is potentially detrimental when walking over ground. One possible consideration is that the threat of the obstacle and the context of the environment caused something similar to a reflex reversal to occur.

2 Objective

The purpose of this exploratory study was to examine adaptations to the elevating response during obstacle tripping while walking on a treadmill. Specifically, we explored the question utilizing two distinct tripping devices. We hypothesized that an open obstacle would influence the elevating response more than a closed obstacle. Specifically, adaptations would be observed in vertical COM range from the time the ipsilateral foot toe off occurs until heel contact and with COM foot height. The current exploratory study provides insight regarding human trip response adaptations developed while perturbed during the early swing phase of the gait cycle. Additionally, the possible transfer effects from adapted responses in the lab to the worksite areexplored.

3 Methods

A quantitative descriptive study design was used to determine if obstacle type influenced the trip recovery strategy during an early swing phase trip. This approach was necessary do to the small sample size that prohibited running a traditional statistical design. However, the data trends allowed for the quantitative descriptive design. The participants were presented with two different types of tripping obstacles (see Fig. 2), open and closed during the early swing phase to elicit the elevating response. As explained in the introduction section, early swing phase trips induce the elevating response. One of two randomly determined obstacles were placed unexpectedly in front of the left limb so that contact was made during early swing phase. The two obstacles were termed closed and open. They were matching in volume (29.5×23.0×7.5 cm) with the mass of the closed obstacle slightly heavier than that of the open (2.8 kg vs. 2.3 kg) (see Fig. 2). The slight difference in weight was determined to be insignificant and mainly due to the extra tape needed for construction. The closed obstacle was created to provide a perturbation similar to a trip when walking over ground and encountering a solid object lying flush on the ground such as a box (Fig. 2). The open obstacle was constructed to “catch” the shoe when contacted, similar to encountering an object that has an open area underneath it or lip around it. For example, when walking on a path outdoors the foot can catch an open box or cord. The Comparative Pain Scale (Harich, 2002) was also used during the tripping trials to determine a threshold for test termination (see Fig. 3). The Comparative Pain Scale (Harich, 2002) consists of a numerical scale where participants rate their pain from zero (no pain) to ten (unimaginable, unspeakable). In general, 1–3 are categorized as mild, 4–6 as moderate, and 7–10 as severe (see Fig. 3). The tripping trials in this study were terminated if a participant reported a value of “3” or higher. No trials were terminated as no participants reported a “3” or higher. Testing occurred in a laboratory setting equipped with a modified treadmill and motion capture system, and lasted approximately 1 1/2 hours per participant. The Institutional Review Board granted ethical approval for the study.

The researchers sought to recruit healthy participants to cover the general working age range and resulted in three age categories, younger (20–30 yrs.), middle-aged (40–50 yrs.), and older (60–70 yrs.). Healthy was defined as being absent of musculoskeletal injury, hypertension, osteoporosis, or uncorrected visual impairments. Recruitment resulted in seven younger (24 ± 3.3 yrs.), four middle-aged (46 ± 3.0 yrs.), and three older (63 ± 3.8 yrs.) male and female participants. Masses and heights were similar between the participants. See Table 1 for participant characteristics.

3.1 Instruments and procedures

After providing written informed consent, anthropometric data were collected and the Comparative Pain Scale was explained to the participants. To reduce the ability to predict when the tripping obstacle would appear, vision (peripheral and lower only) and hearing were occluded using goggles and noise canceling headphones playing white noise, respectively (see Fig. 4). To prevent falling to the ground (treadmill belt), a harness that was attached at one end to the ceiling, and placed loose around the body so that it provided enough slack for free motion, but also restraint in the event of an impending fall. Once prepared, participants were then instructed to walk on a modified treadmill (front panel and side rails removed; see Fig. 5). Treadmill speed was the preferred speed determined by the participant. Participants walked on the treadmill until acclimated. Then treadmill speed was increased and decreased until the participant informed the researcher when a comfortable speed was reached. The researcher would then increase and decrease the speed of the treadmill above and below the comfortable walking speed multiple times (minimum of three). Increasing and decreasing treadmill speed continued until three consecutive identically reported comfortable walking speeds were reached [24].

3.2 Data collection and analysis

Gait kinematics were recorded using three Optotrak cameras (Northern Digital Inc., Ontario, Canada) sampling at 100 Hz. They captured the motion of 18 infrared emitting diodes that were attached bilaterally to the major joints of each participant and used to define 12 body segments. All participants were tripped on their left limb during early swing phase while unconstrained on the modified treadmill. To ensure that all trips occurred during early swing phase a percentage from each participant’s entire swing phase sagittal horizontal displacement was calculated (1–33% early, 34–66% mid, and 67–100% late). Five strides from trial one were averaged to determine the average step length of the left foot. Leg dominance was not determined. Each obstacle, in random order, then perturbed participants during early swing phase 20 times.

Trip recovery outcomes were quantified using kinematic measures that have previously been shown to differ between successful and failed trip recoveries [20] and primarily included measures related to trunk control in order to prevent a fall; specifically, the vertical displacement of whole body center of mass (COM) (cm) and the foot height deviation (cm). The location of the center of mass (COM) for each segment was determined using Winter’s [25] method (2005) based upon a percentage of each body segment length. Once the location of each body segment COM was determined the sum of their coordinates approximated the total body COM location [25]. All analyses were conducted using a custom MATLAB^® (Mathworks, Natick, MA, USA) program. The current exploratory study focused primarily on the elevating response adaptations.

3.3 Overall analysis

Kinematic data were examined for mean differences between the three age categories and between the two obstacles. We hypothesized that the open obstacle would influence the elevating response more than the closed obstacle. Specifically, adaptations would be observed in vertical COM range from the time the ipsilateral foot toe off occurs until heel contact and COM foot height. Additionally, we hypothesized that obstacle context would influence the recovery response of the older adults more than the other age categories.

4 Results

4.1 Participant characteristics

Participant characteristics for the age categories, younger, middle-aged, and older adults are presented in Table 1. Participants in each age category did, incur “missed” trials. A missed trial meant the foot did not make contact with the obstacle. An average of 4-5 trials were missed by each participant in each age category.

4.2 Recovery strategy adaptation

All trip induced trials occurred during early swing phase (first 33% of the swing phase). However, in addition to the elevating strategy, a novel “push” strategy was observed. Participant “initially lifted and then kicked” the obstacle out of the way with the recovery foot during this strategy. This occurred across all age categories (≈79%) but with increased frequency in the older adults and with the open obstacle. The mean number of trials resulting in a push were 1.3 ± 1.4, 7.5 ± 3.6, and 9.0 ± 4.6 for the younger, middle-aged, and older adults, respectively.

4.3 Vertical displacement of whole body center of mass

Figure 6 depicts the vertical center of mass range for the closed, open, and push responses, respectively. Older adults exhibited the largest vertical center of mass range for the open (0.96 cm) and closed obstacles (1.02 cm) as well as responses resulting in a push (0.72 cm). Thus, largely the obstacles and the push response, compared to younger and middle-aged participants, influenced the center of mass vertical height the most for older participants.

4.4 Foot height deviation

Figure 7 shows the foot height deviation of the three age categories for the closed, open, and push response. The middle-aged participants produced the highest foot height for both obstacle conditions and for trials where a push response was observed. Across the three age categories, the open obstacle condition exhibited the highest mean foot heights.

5 Conclusions

This exploratory study examined gait recovery from an early swing phase trip utilizing two obstacles, one open to catch the foot and one closed. In addition, the elevating recovery response strategy was explored among a large age range, young, middle-aged and older adults. Information obtained from the current study maybe useful for establishing guidelines for developing fall prevention programs (i.e. functional fall training) in the workplace. Overall, findings suggest that obstacle type and age influence the trip recovery response. Successful recovery from a trip involves reducing the rate of forward movement of the body and having proper placement of the recovery foot. The open and closed obstacle features not only influenced the recovery strategy but an entirely new response was displayed, termed “push.” Invariant features of the obstacle may have been rapidly acquired by participants providing information such as “movability” and thus allowed for initial contact and then push of the obstacle out of the way torecover.

The push response was unique in that once the foot caught on an obstacle during an early swing phase trip, participants recovered with a push by lifting the foot and then “kicking” the obstacle out of the way. An obstacle being caught and lifted by the shoe has been documented in trip research literature; however, the obstacle returned to the treadmill and no linear translation was reported. The only result reported was a higher maximum foot height [26].

The possible increased threat caused by the obstacles in the current study resulted in some participants pushing the obstacle away from the body. Pushing was done instead of lifting the obstacle with the shoe until the obstacle fell off the foot and thus became clearable. The push response is in agreement with cognitive influences affecting movement strategies [9]. Furthermore, the push response demonstrated that movement strategies are not stereotyped and perceived threat influences recovery responses [9]. However, compared to obstacles used in prior works that may have influenced response strategies, chance of injury did not increase in the current study due to the more compliant materials used to make the obstacles [12, 26].

The push response enabled a successful recovery by causing the obstacle to move away from the participant, whereas the obstacle used by Eng [9] folded back and lied flush with the ground. We infer that the frequency of a push response would increase if the obstacle were not attached to the ground or by a hinge mechanism, as in prior research [9, 12]. A compliant obstacle, such as the one used by Eng [9], but unhinged from the ground, would perhaps provide the participant with a response option of moving the obstacle away. Limiting movability of the obstacle possibly reduces recovery options.

Older adults exhibited the largest vertical center of mass range for the open and closed obstacles as well as responses resulting in a push. Thus, largely the obstacles and the push response influenced the center of mass vertical height the most for older adults compared to younger and middle-aged participants. Interestingly, the older adults had the largest whole body COM vertical fluctuations during their recovery for both obstacles. Increased vertical COM range for the obstacles and recovery responses resulting in a push suggests that the older adults were less able to adapt after the trip, when compared to the other two age categories. Future studies should examine the training effect to determine if multiple trip sessions will decrease this range. Across the three age categories, the open obstacle condition exhibited the highest mean foot heights. The middle-aged participants produced the highest foot height for both obstacle conditions and for trials resulting in a push response.

Some studies that have used a treadmill to induce trips may be less ecologically relevant [27]. These studies perturbed the participant by rapidly accelerating a treadmill while the participant is standing still. However, workers do not fall when standing still and the ground does not move suddenly from under their feet. There is minimal hip, knee, or trunk flexion and extension occurring while standing. Therefore, standing does not exhibit the elevating and lowering responses that would occur at a worksite. To trip, gait needs to have been initiated. We suggest that fall training for the workplace utilizing a treadmill should have participants walking prior to the onset of a perturbation, such as a sudden increase in acceleration. Furthermore, fall research that utilizes trips may benefit from adding an obstacle on the treadmill to increase the threat of a fall after the perturbation. Trip research may benefit from obstructions occurring during the gait cycle, and doing so could potentially increase recovery response options to prevent a fall; thus, possibly decreasing falls from trips in the workplace. Future research could address this theory.

In summary, the factors that influence whether a trip recovery will be successful are multifactorial with both intrinsic (i.e. history of falling, balance, advanced age, impaired strength, poor reaction time) and extrinsic (i.e. poor lighting, sidewalk cracks, cords and wires on the floor, loose carpeting) factors that have been identified. Current preventative strategies for the workplace focus on factors such as facility design [15] instead of functional training. Although hazards are reduced, workers will still trip and thus possibly fall. Trip training may help reduce falls in the workplace by increasing the ability to recover after a perturbation. Furthermore, fall prevention programs targeting improving the intrinsic factors such as strength and balance and have resulted in inconsistent effectiveness [17].

The current study utilized obstacles varying in physical properties that moved with the treadmill to trip participants of varying ages during the early swing phase of the gait cycle. The most common corrective response was the elevating response. The second most common reaction was the push response, meaning kicking the obstacle out of the way, which all but three participants exhibited at some point during testing. The push response adaptation after early swing phase perturbations increased the number of behavioral response strategies. Previous research has reported only the elevating response[9 , 14].

Financial ramification from unrecovered trips resulting in a fall will continue to occur in the workplace. Therefore, finding effective and efficient ways to improve the recovery response following trips is crucial. The current study provided support that obstacle type influences the behavioral response after a trip, specifically an obstacle that catches the shoe. Therefore, these types of obstacles should be considered when designing functional fall programs. Furthermore, information gained from the current study could be useful if establishing guidelines for the development of a fall prevention program in the workplace. Based on the results of this exploratory study, adding open type obstacles to a fall training program to mimic trips from such items as open boxes, cords, and branches could be beneficial. However, these assumptions as well as limitations in this study deserve discussion.

Future trip research should examine the long-term effects of multiple obstacles and training. Additionally, future research should examine transfer between limbs and environments, specifically outside the lab. It is unclear if the results from the current study are retained and transfer to an environment outside the laboratory. An increased sample size would have improved the strength of the current study by allowing the use of traditional statistical analysis. Additionally, the variables used to determine adaptation (trunk angle versus vertical toe height) may not represent an improved ability to recover from a trip. Future research examining the relationship between these two variables would need to be quantified utilizing techniques such as vector coding. More research examining interlimb and environmental transfer can improve our understanding of rapid improvements observed during repeated exposure to a perturbation.

Conflict of interest

None to declare.

References

Bureau of Labor Statistic-United States Department of Labor. National Census of fatal occupational injuries summary 2013; http://www.bls.gov/news.release/cfoi.nr0.htm

Stevens

, Corso

, Finkelstein

, Miller

. The cost of fatal and non-fatal falls among older adults. Injury Prevention 2006;12:290–295.

Lipscomb

, Glazner

, Bondy

, Guarini

, Lezotte

. Injuries from slips and trips in construction. Applied Ergonomics 2006;37:267–274.

Schiller

, Kramarow

, Dey

. Fall injury episodes among noninstitutionalized older adults: United States, 2001-2003. Advanced Data 2007;21:1–16.

Alexander

, Rivara

, Wolf

. The cost and frequency of hospitalization for fall-related injuries to older adults. American Journal of Public Health 1992;82:1020–1023.

Avineri

, Shinar

, Susilo

. Pedestrians’ behaviour in cross walks: The effects of fear of falling and age. Accident Analysis & Prevention 2012;44:30–34.

Dutta

. Significance of sarcopenia in the elderly. Journal of Nutrition 1997;127:992–993.

Maki

. Gait changes in older adults: Predictors of falls or indicators of fear. Journal of the American Geriatrics Society 1997;45:313–320.

Eng

, Winter

, Patla

. Strategies for recovery from a trip in early and late swing during human walking. Experimental Brain Research 1994;102:339–349.

10.

Eng

, Winter

, Patla

. Intralimb dynamics simplify reactive control strategies during locomotion. Journal of Biomechanics 1997;30:581–588.

11.

Crenshaw

, Kaufman

, Grabiner

. Trip recoveries of people with unilateral, transfemoral or knee disarticulation amputations: Initial findings. Gait and Posture 2013;38:534–536.

12.

Pijnappels

, Bobbert

, van Dieen

. Contribution of the support limb in control of angular momentum after tripping. Journal of Biomechanics 2004;37:1811–1818.

13.

Dietz

. Human neural control of automatic functional movements: Interaction between central programs and afferent input. Physiological Reviews 1992;72:33–69.

14.

Schillings

, van Wenzel

BMH

, Mulder

, Duysens

. Muscular responses and movement strategies during stumbling over obstacles. Journal of Neurophysiology 2000;94:1158–1168.

15.

Maynard

. Prevention through design-slips, trips and falls. In American Society of Safety Engineers-a Technical Publication of ASSE’s Ergonomics Practice Specialty 2014;4:1–19.

16.

Aronovitch

. Falls and patients safety for older adults. Ostomy Wound Management 2006;52:22–28.

17.

Wright

, Hyner

. Principal barriers to health promotion program participation by older adults. American Journal of Health Studies 2008;24:215–222.

18.

Kerse

, Butler

, Robinson

, Todd

. Fall prevention in residential care: A cluster, randomized, controlled trial. Journal of the American Geriatrics Society 2004;52:524–531.

19.

Brauer

, Burns

, Galley

. A prospective study of laboratory and clinical measures of postural stability to predict community-dwelling fallers. Journal of Gerontology 2000;55:469–476.

20.

Pavol

, Owings

, Foley

, Grabiner

. Mechanisms leading to a fall from an induced trip in healthy older adults. Journal of Gerontology 2001;56:428–437.

21.

Bhatt

, Wang

, Pai

. Retention of adaptive control over varying intervals: Prevention of slip induced backwards balance loss during gait. Journal of Neurophysiology 2006;95:2913–2922.

22.

Bieryla

, Madigan

, Nussbaum

. Practicing recovery from a simulated trip improves recovery kinematics after an actual trip. Gait and Posture 2007;26:208–213.

23.

Grabiner

, Donovan

, Bareither

, Marone

, Hamstra-Wright

, Gatts

, Troy

. Trunk kinematics and fall risk of older adults: Translating biomechanical results to the clinic. Journal of Electromyography and Kinesiology 2008;18:197–204.

24.

Holt

, Hamill

, Andres

. Predicting the minimal energy costs of human walking. Medicine and Science in Sports and Exercise 1991;23:491–498.

25.

Winter DA

. Biomechanics and motor control of human movement (3rd ed.). John Wiley & Sons, Inc., Hoboken, New Jersey; 2005.

26.

Schillings

, van Wenzel

BMH

, Duysens

. Mechanically induced stumbling during treadmill walking. Journal of Neuroscience Methods 1996;67:11–17.

27.

Owings

, Pavol

, Grabiner

. Mechanisms of failed recovery following postural perturbations on a motorized treadmill mimic those associated with an actual forward trip. Clinical Biomechanics 2001;16:813–819.