People with Parkinson disease (PD) typically demonstrate low amplitude, or hypokinetic, movements.1 These stereotypical movements occur during fine motor tasks such as handwriting,2 as well as gross motor tasks like walking.3 In addition to these hypokinetic, planned movements, people with PD also demonstrate underscaled reactive movements such as stepping in response to a balance perturbation.4 These poor postural responses are associated with increased fall risk5 and occur in the forward, backward, and lateral directions.6–8 Although reactive stepping responses have been well studied, to our knowledge, investigators have not yet examined the capacity of individuals with PD to produce a large, single step in a voluntary manner.
The Maximum Step Length Test (MSLT) was first described by Medell and Alexander.9 It involves an individual stepping as far as possible in a given direction (eg, forward, backward, and lateral) and returning to a standing position without a loss of balance. Performance of the MSLT has been studied in older adults. MSLT performance is related to measures of balance such as the Berg Balance Scale, Functional Reach Test, and Tinetti Performance Oriented Mobility Assessment in older adults.10,11 Furthermore, Cho et al11 showed that the MSLT accurately identifies older adults who are at risk for future falls. From a clinical perspective, an added advantage of the MSLT is that it can be done quickly and requires only a measuring tape (demarcated in centimeters) to administer. Given the MSLT is feasible for clinical use and can be used to assess movements that are likely difficult for people with PD, we investigated MSLT performance in this population.
The purpose of this study was three-fold. We aimed to (1) characterize MSLT performance in the forward, backward, and lateral directions in a sample of individuals with PD, (2) determine the effect of anti-PD medication on MSLT performance, and (3) investigate the construct validity of the MSLT by assessing relationships between the MSLT and measures of PD motor severity, balance, balance confidence, functional mobility, and gait. We had several hypotheses. First, MSLT distances would be shorter in the backward direction than in forward, given that people with PD tend to have more difficulty with movement in the backward direction.12 Second, MSLT performance would be worse in the OFF-medication condition than in the ON-medication condition in people with PD because movement is known to be more difficult in PD without dopamine replacement therapy.13 Third, MSLT performance in all directions would be related to measures of balance, balance confidence, gait, and PD motor sign severity, given that the movement inherent to the MSLT incorporates elements of balance and gait.9,11
This study was conducted as a cross-sectional study within a large exercise trial (NCT01768832).14 Participants included in this study were a sample of convenience from the larger trial and were consecutively recruited to perform the MSLT during their scheduled laboratory visit. To be included in the exercise trial, participants must have been (1) diagnosed with idiopathic PD, (2) at least 30 years of age, (3) between Hoehn and Yahr (H&Y) stages I and IV, and (4) able to walk independently with or without an assistive device for at least 3 m. Participants were excluded if they had a Mini-Mental Status Examination score less than 24, a neurologic condition other than PD (eg, stroke), and undergone surgical management for PD. All participants provided written informed consent in accordance with the policies and procedures of the Human Research Protection Office at Washington University in St Louis.
Maximum Step Length Test
Each participant performed the MSLT in 3 stepping directions: (1) forward, (2) backward, and (3) lateral. This order was the same for each participant for both medication conditions. For the MSLT, the participant wore shoes and stood with the arms folded across the chest. The participant was instructed to step out as far as possible and then return to the original standing position. For each stepping direction, 3 practice trials were given before 5 recorded test trials with each leg. Trials in which an error occurred were discarded. Errors were defined as one or more of the following occurring during a trial: unfolding of the arms, displacement of the stance limb, inability to return to the starting stance position, noncompliance with stepping direction, or a loss of balance requiring physical assistance to regain stability.9,11 For each stepping direction, participants stepped with both dominant and nondominant legs. Leg dominance was determined by asking each participant, “If a ball were rolled directly at you, with which foot would you kick the ball?” If a participant responded that they would kick the ball with the right leg, the right leg was deemed the dominant leg.15
For the forward and backward stepping directions, the starting position was with the feet approximately shoulder-width apart. The most anterior portion (ie, the toe portion of the shoe) of the stepping foot was centered on the measuring tape at zero. The measuring tape was marked in centimeters. The measurement was collected from the point where the most anterior portion (ie, the toe portion of the shoe) of the stepping foot landed. For the lateral directions, the starting position was with the feet close together but not touching. This was done to maximize each participant's ability to step in the lateral direction. The midpoint (ie, middle of the shoe when viewing the participant head-on) of the stepping foot was centered on the measuring tape at zero. The measurement in the lateral direction was collected from the point where the midpoint of the stepping foot landed. All measurements, to the nearest centimeter, were collected using visual observation of the examiner, who immediately recorded each trial on the data collection form. The examiner was not blinded to medication status. A research assistant stood near the participant to provide assistance for safety if needed.
MSLT measures were normalized to leg length to account for the effect of leg length on MSLT performance. This was done by dividing the average MSLT stepping distance for each direction (cm) by the leg length of the stepping leg (cm). The MSLT has been reported to have excellent inter- and intrarater reliability (intraclass correlation coefficient [ICC] = 0.96 and ICC = 0.95, respectively) in older adults.10 Given the correlation between the MSLT and single-limb stance time (r = 0.68), Timed Up and Go (TUG) (r = −0.65), and functional reach test (r = 0.65), the MSLT has been reported to be a valid measure of balance performance in older adults.10
Motor Performance Tests
To assess construct validity of the MSLT, we utilized measures of balance, balance confidence, functional mobility, gait, and PD motor severity. Balance was measured using the Mini-Balance Evaluation Systems Test (Mini-BESTest).16 This test includes 14 items, each scored on a rank scale of 0 to 2, with 2 indicating no impairment in balance. Higher scores indicate better balance. There are 4 subsections of the Mini-BESTest, which assess the following unique balance systems: (1) anticipatory postural adjustments, (2) reactive postural responses, (3) sensory orientation, and (4) stability in gait. For the purposes of this study, we collected the total and individual subsection scores. Self-perceived balance confidence was assessed using the Activities-specific Balance Confidence (ABC) scale.17 This 16-item questionnaire addresses a participant's perceived balance confidence when performing specific tasks. A score of 0 indicates no confidence in balance during a particular task, whereas 100% indicates complete confidence in balance during a particular task. The average score of the 16 items was calculated. Functional mobility was measured using the TUG18 (seconds), which was collected as part of item 14 within the Mini-BESTest.
Gait velocity was measured using an instrumented 5-m walkway (GAITRite, CIR System Inc., Franklin, NJ) to acquire data related to spatiotemporal gait parameters. Participants performed 3 trials at their self-selected forward walking speed. The average velocity (meters/second) was normalized to leg length. Walking endurance was measured using the 6-minute walk test (6MWT).19 Participants were instructed to “cover as much ground as possible” over 6 minutes. The distance (meters) traversed over 6 minutes was recorded.
Motor sign severity was measured using the Movement Disorder Society-Unified Parkinson Disease Rating Scale subsection III (MDS-UPDRS III).20 This 33-item scale is a gold standard for assessment of bradykinesia, rigidity, tremor, postural instability, and gait difficulty in persons with PD.
Participants completed the full test battery in both the OFF- and ON-medication conditions in the following order: (1) MDS-UPDRS III, (2) Mini-BESTest, (3) GAITRite, and (4) MSLT. The tests were done in a fixed order given the protocol of the original clinical trial. Because the MSLT was not part of the protocol for the original clinical trial, it was completed last to avoid impacting performance of any other measures. The MDS-UPDRS III and Mini-BESTest were scored by a rater trained in administering and rating these assessments. The rater was blinded to medication status. The ABC scale was completed before each participant's laboratory visit. OFF-medication was defined as greater than or equal to 12 hours since last intake of anti-PD medication. ON-medication evaluations began at least 45 minutes and no more than 1.5 hours after intake of anti-PD medication.
Descriptive statistics were used to characterize the sample. Normality was assessed using Shapiro-Wilk tests. A 2-factor repeated-measures analysis of variance was used to assess for main effects of medication state (“OFF” vs “ON”) and stepping direction (forward vs backward vs lateral), as well as interactions between medication state and stepping direction (α = 0.05). Post hoc tests were conducted as appropriate. The relationships between MSLT and motor performance tests for OFF- and ON-medication conditions were assessed using Pearson or Spearman correlations (α = 0.05) as appropriate. OFF-medication MSLT stepping distances were used in correlations run for physical performance measures collected OFF-medication. Similarly, ON-medication MSLT stepping distances were used in correlations run for physical performance measures collected ON-medication. Correlations were run between the ABC score and both OFF- and ON-medication MSLT stepping distances.
Of 49 eligible participants, 40 participants with mild to moderate PD completed testing in both OFF- and ON-medication conditions (Table 1). Nine participants did not perform the MSLT for the following reasons: not on levodopa (n = 5), excessive fatigue (n = 3), and hip pain (n = 1). Of the 9 participants not performing the MSLT, 7 were H&Y II with OFF-medication state MDS-UPDRS III scores ranging from 26 to 43, indicating their motor sign severity was similar to the sample completing the MSLT. The additional 2 participants not performing the MSLT, both due to excessive fatigue, were H&Y III with OFF-medication state MDS-UPDRS III scores of 50 and 52, respectively. There were no adverse events associated with MSLT performance.
Table 1 -
Participant Demographics (N = 40)
|Gender (female), n (%)
||65.1 ± 8.2
|Years with PDa
||4.8 ± 3.7
|Levodopa equivalent daily dose, mga
||898.4 ± 601.1
|Hoehn and Yahr, stage (n)
||I (1), II (34), III (5)
|OFF MDS-UPDRS IIIb
|ON MDS-UPDRS IIIb
|OFF TUG, sa
||10.5 ± 2.7
|ON TUG, sa
||9.7 ± 2.4
|OFF 6MWT, ma
||449.6 ± 99.1
|ON 6MWT, ma
||483.6 ± 105.3
|OFF gait velocity, m/sa
||1.4 ± 0.2
|ON gait velocity, m/sa
||1.5 ± 0.3
Abbreviations: ABC, Activities-specific Balance Confidence; MDS-UPDRS, Movement Disorder Society-Unified Parkinson Disease Rating Scale; Mini-BESTest, Mini-Balance Evaluation Systems Test; PD, Parkinson disease; 6MWT, 6-minute walk test; TUG, Timed Up and Go.
aValues are mean ± standard deviation.
bValues are median (interquartile range).
All MSLT data were normally distributed. Although there were statistically significant differences in MSLT stepping distance between dominant and nondominant legs, the mean raw data indicate that these differences were small (ie, ≤3.71 cm regardless of medication status for each direction; see Supplemental Digital Content 2, Table 1, http://links.lww.com/JNPT/A187). Furthermore, the relationships between motor performance measures and MSLT dominant and nondominant stepping distances were similar (see Supplemental Digital Content 2, Tables 2 and 3, http://links.lww.com/JNPT/A187). As such, for all analyses, we used the average stepping distance (overall mean, collapsing across dominant and nondominant legs) normalized to leg length.
There was a significant effect of direction ((F(2,38) = 23.86, P < 0.001) such that participants stepped further in the forward direction when compared with the backward (t = 8.95, P < 0.001) and lateral (t = 2.58, P = 0.012) directions (Figure). Participants also stepped further in the lateral direction than they did in the backward direction (t = −5.95, P < 0.001). There was a significant effect of medication state on MSLT performance such that participants had increased stepping distance when ON-medication compared with OFF-medication, regardless of stepping direction (F(1,39) = 9.52, P = 0.004) (Figure). There was no direction by medication interaction (P = 0.44).
In the OFF-medication state, all MSLT directions were strongly related to each other (r ≥ 0.89, P < 0.001) (Table 2). The strength of correlations between all MSLT distances, regardless of direction, and measures of motor performance ranged from weak to strong (range of r: −0.38 to 0.81) (Table 3). The MSLT was most strongly related to the Mini-BESTest (r ≥ 0.72, P < 0.001) and 6MWT (r ≥ 0.73, P < 0.001). In the ON-medication state, all MSLT directions were strongly related to each other (r ≥ 0.86, P < 0.001) (Table 2). Except for the relationships between MSLT forward and lateral with MDS-UPDRS III as well as MSLT lateral with ABC, all MSLT distances were significantly related to all measures of motor performance (range of r: −0.26 to 0.76), with the strongest relationship between MSLT and 6MWT (r ≥ 0.75, P < 0.001) (Table 3).
Table 2 -
Correlations (Pearson) Between MSLT Directions
Abbreviation: MSLT, Maximum Step Length Test.
aP ≤ 0.01.
bP < 0.05.
Table 3 -
Correlations Between MSLT and Motor Performance
Abbreviations: ABC, Activities-specific Balance Confidence; APA, anticipatory postural adjustments; MDS-UPDRS, Movement Disorder Society-Unified Parkinson Disease Rating Scale; Mini-BESTest, Mini-Balance Evaluation Systems Test; MSLT, Maximum Step Length Test; PD, Parkinson disease; PR, postural responses; 6MWT, 6-minute walk test; SG, stability in gait; SO, sensory orientation; TUG, Timed Up and Go.
bP ≤ 0.01.
cP < 0.05.
To our knowledge, this is the first study of MSLT performance in people with PD. According to our results, regardless of medication state, forward MSLT performance is greater than both backward and lateral directions. This finding is not surprising, given that people with PD are known to have difficulty with postural responses21,22 and gait23,24 in the backward direction.12 Our findings extend this previous work by demonstrating that people with PD have greater voluntary stepping hypokinesia in the backward direction when compared with forward. Interestingly, Medell and Alexander9 reported no directional differences in MSLT performance in unimpaired young, unimpaired older, and impaired older women. This further supports the notion that difficulty moving in the backward direction distinguishes people with PD from neurologically healthy individuals. It is possible that the poorer backward MSLT is due to hypokinesia and to poorer and more variable anticipatory control25 in persons with PD. With respect to raw stepping distance, the sample with PD had greater stepping distances (ranging from 70.82 to 84.96 cm—see Supplemental Digital Content 2, Table 1, http://links.lww.com/JNPT/A187), regardless of direction, compared with impaired older adults (ie, ≥2 falls in the past year and complaints of frequent unsteadiness; approximately 65 cm), but lower stepping distances than unimpaired older adults (approximately 90 cm).9 The lack of postural instability in our sample of participants with PD (ie, 35 of 40 participants at H&Y II or less) may explain their larger stepping distance compared with impaired older adults. However, these comparisons should be interpreted with caution as Medell and Alexander9 included only women who used only the right leg to perform the MSLT.
Biomechanical factors cannot be ruled out as having unique contributions to MSLT performance in the PD population. For example, stooped posture, which is common in PD,13 may disproportionately shift the center of mass anteriorly, thus making backward movement more difficult. Lower extremity muscle rigidity,13 particularly in the hip flexors, may limit the ability to extend the hips during the backward stepping movement. Finally, muscle strength,26,27 if impaired in the stepping or stance limb, may negatively impact postural control during MSLT performance. Investigators should measure these biomechanical factors in the future to determine how they impact performance of each MSLT direction in PD.
We also show that people with PD demonstrate a statistically significant improvement in MSLT performance when ON-medication compared with OFF-medication. However, the magnitude of the differences between OFF- and ON-medication MSLT performance (normalized to leg length) is small (Figure and Supplemental Digital Content 2, Table 1, http://links.lww.com/JNPT/A187) and may not be clinically meaningful as Goldberg et al10 have reported the MSLT to have a minimal detectable change of 7.32 inches (18.6 cm) in older adults. None of the differences between OFF- and ON-medication conditions exceed 5 cm (Supplemental Digital Content 2, Table 1, http://links.lww.com/JNPT/A187), suggesting that these differences may not reflect a meaningful change. Investigators have reported that improved gait velocity with levodopa is driven by an increase in step length, suggesting levodopa uniquely impacts movement size.28–30 Although our results fail to show a large increase in MSLT stepping distance when ON-medication compared with OFF-medication, the MSLT should not be considered solely as a measure of movement amplitude capacity. The MSLT involves many elements of postural control during performance of the test (eg, anticipatory weight shift, single-limb stance, and maintaining balance within a new base of support); the strong relationship between the MSLT and Mini-BESTest is evidence of these requirements. The effects of levodopa on postural control and balance are less clear. Although investigators have reported improved Mini-BESTest31 and Functional Gait Assessment32 scores ON-medication compared with OFF, others have noted worsened sway during static stance with levodopa.33,34 Our sample demonstrated only a 1-point increase in the Mini-BESTest score ON-compared with ON-medication, which is similar to the small magnitude “OFF” to “ON” change in MSLT performance. As such, if the MSLT is assumed to be reflective of balance in PD, our results are in keeping with those studies reporting no clinically meaningful improvement in postural control when on levodopa.
Within the MSLT, all directions were strongly related to one another regardless of medication status. The positive, strong correlations suggest that those who step farther in the forward direction also step farther in the backward and lateral directions. This suggests that the underlying construct measured by the MSLT is similar regardless of direction, and calls into question whether the information gained from each direction is unique. However, the fact that people step farther forward than they do backward or to the side suggests that although the underlying construct measured by the MSLT may be similar regardless of direction, other factors (eg, biomechanical factors) might account for the observed differences in stepping length across directions. Investigators should continue to study the properties of the MSLT to determine whether each of the 3 directions provides unique clinical information and determine how biomechanical factors impact MSLT performance.
Regarding the relationships between MSLT and measures of motor performance, the strongest relationships exist between MSLT (regardless of direction) and the Mini-BESTest, TUG, and 6MWT, respectively. The relationships between the MSLT and Mini-BESTest and TUG are not surprising, given that assessment of anticipatory postural control and stability in gait is included within the Mini-BESTest16 and TUG.18 These constructs seem to overlap with the MSLT; however, the unexplained variance in the relationships suggests that the MSLT is measuring something not captured within the Mini-BESTest and TUG, particularly when assessed ON-medication. Interestingly, except for the stability in gait subsection in the ON-medication condition, all of the correlations between the individual Mini-BESTest subsections and MSLT directions were weaker than those between the Mini-BESTest total score and MSLT directions. As previously noted, the MSLT requires multiple balance components including anticipatory postural adjustment, single-limb balance with concurrent movement of the opposite lower extremity, and static balance control in-stride with an extended base of support. It is possible that the correlations between the individual subsections of the Mini-BESTest were weaker because they capture performance of a singular construct, whereas the MSLT likely captures performance spanning multiple constructs. The similarities between the correlations of the stability in gait subsection with MSLT and Mini-BESTest total with MSLT potentially suggest this subsection and the MSLT are measuring similar constructs in the ON-medication state. However, the relationships in the OFF-medication state do not confirm this. The relationship between MSLT performance and measures of balance has been reported in the older adult population, and the strength of these relationships is similar to that of relationships reported herein.10,11 The strong relationship between the MSLT and the 6MWT, regardless of medication status, was not expected. However, Falvo and Earhart35 reported that 6MWT distance is partly explained by impaired balance in PD, whereas Canning et al36 reported that gait hypokinesia impacts 6MWT performance. Taken together, this suggests that the relationship between the 6MWT and MSLT may be explained by the constructs of balance and gait hypokinesia in the PD population.
This study is not without limitations. First, this was a relatively small sample of people with PD with mild to moderate motor severity. To this point, 87.5% of the study sample was classified as H&Y I or II, indicating no postural instability as measured by the MDS-UPDRS III.20 As such, the results may not be generalizable to the PD subpopulation with significant postural instability, creating the need for validation of these findings in a sample with PD exhibiting a wider range of motor sign severity. Related to the procedures, OFF-medication testing was always completed before ON-medication testing, so we cannot rule out that a learning effect may have contributed to the results. This testing order was chosen to minimize participant burden by avoiding 2 separate laboratory visits. Also, the examiner was not blinded to medication status; however, the small magnitude of differences between OFF- and ON-medication performance suggests that potential bias is unlikely to have influenced the results. Future studies should randomize the order of testing, blind the examiner, and attempt to study OFF- and ON-medication MSLT performance on separate days. Because this was a cross-sectional study and reflects participant performance at only one point in time, investigators should also study the variability in MSLT performance on different days within the same medication state allowing for calculation of the minimal detectable change, this will facilitate determination of the stability of the MSLT over time in PD. Finally, we lacked a sufficient number of participants with a fall history, prohibiting further analyses. However, these results warrant further investigation into the ability of the MSLT to predict fall risk in people with PD.
In people with PD, stepping distances for the MSLT were greatest in the forward direction followed by lateral and backward stepping. They demonstrated improved MSLT performance, albeit small in magnitude, in the ON-medication state compared with the OFF-medication state. Irrespective of medication status, the MSLT has moderate to strong relationships with the Mini-BESTest, TUG, and 6MWT. Overall, MSLT performance appears to be explained by balance and gait hypokinesia in people with PD.
Thank you to Ellen Sutter for assistance with data organization and analysis as well as Rich Nagel and Martha Hessler for assistance with data collection.
1. Morris ME, Iansek R, Matyas TA, Summers JJ. The pathogenesis of gait hypokinesia in Parkinson's disease. Brain. 1994;117(pt 5):1169–1181.
2. Wagle Shukla A, Ounpraseuth S, Okun MS, Gray V, Schwankhaus J, Metzer WS. Micrographia and related deficits in Parkinson's disease: a cross-sectional study. BMJ Open. 2012;2(3).
3. Mak MK. Reduced step length, not step length variability is central to gait hypokinesia in people with Parkinson's disease. Clin Neurol Neurosurg. 2013;115(5):587–590.
4. Jacobs JV, Horak FB. Abnormal proprioceptive-motor integration contributes to hypometric postural responses of subjects with Parkinson's disease. Neuroscience. 2006;141(2):999–1009.
5. Jacobs JV, Earhart GM, McNeely ME. Can postural instability tests improve the prediction of future falls in people with Parkinson's disease beyond knowing existing fall history? J Neurol. 2016;263(1):133–139.
6. King LA, St George RJ, Carlson-Kuhta P, Nutt JG, Horak FB. Preparation for compensatory forward stepping in Parkinson's disease. Arch Phys Med Rehabil. 2010;91(9):1332–1338.
7. de Kam D, Nonnekes J, Oude Nijhuis LB, Geurts AC, Bloem BR, Weerdesteyn V. Dopaminergic medication does not improve stepping responses following backward and forward balance perturbations in patients with Parkinson's disease. J Neurol. 2014;261(12):2330–2337.
8. King LA, Horak FB. Lateral stepping for postural correction in Parkinson's disease. Arch Phys Med Rehabil. 2008;89(3):492–499.
9. Medell JL, Alexander NB. A clinical measure of maximal and rapid stepping in older women. J Gerontol A Biol Sci Med Sci. 2000;55(8):M429–M433.
10. Goldberg A, Schepens S, Wallace M. Concurrent validity and reliability of the maximum step length test in older adults. J Geriatr Phys Ther. 2010;33(3):122–127.
11. Cho BL, Scarpace D, Alexander NB. Tests of stepping as indicators of mobility, balance, and fall risk in balance-impaired older adults. J Am Geriatr Soc. 2004;52(7):1168–1173.
12. Bryant MS, Rintala DH, Hou JG, Lai EC, Protas EJ. Effects of levodopa on forward and backward gait patterns in persons with Parkinson's disease. NeuroRehabilitation. 2011;29(3):247–252.
13. Fahn S. Description of Parkinson's disease as a clinical syndrome. Ann N Y Acad Sci. 2003;991:1–14.
14. Earhart GM, Duncan RP, Huang JL, Perlmutter JS, Pickett KA. Comparing interventions and exploring neural mechanisms of exercise in Parkinson disease: a study protocol for a randomized controlled trial. BMC Neurol. 2015;15:9.
15. Brophy R, Silvers HJ, Gonzales T, Mandelbaum BR. Gender influences: the role of leg dominance in ACL injury among soccer players. Br J Sports Med. 2010;44(10):694–697.
16. Franchignoni F, Horak F, Godi M, Nardone A, Giordano A. Using psychometric techniques to improve the Balance Evaluation Systems Test: the mini-BESTest. J Rehabil Med. 2010;42(4):323–331.
17. Powell LE, Myers AM. The Activities-specific Balance Confidence (ABC) Scale. J Gerontol A Biol Sci Med Sci. 1995;50A(1):M28–M34.
18. Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39(2):142–148.
19. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med. 2002;166(1):111–117.
20. Goetz CG, Tilley BC, Shaftman SR, et al. Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS): scale presentation and clinimetric testing results. Mov Disord. 2008;23(15):2129–2170.
21. Di Giulio I, St George RJ, Kalliolia E, Peters AL, Limousin P, Day BL. Maintaining balance against force perturbations: impaired mechanisms unresponsive to levodopa in Parkinson's disease. J Neurophysiol. 2016;116(2):493–502.
22. McVey MA, Amundsen S, Barnds A, et al. The effect of moderate Parkinson's disease on compensatory backwards stepping. Gait Posture. 2013;38(4):800–805.
23. Hackney ME, Earhart GM. Backward walking in Parkinson's disease. Mov Disord. 2009;24(2):218–223.
24. Tseng IJ, Jeng C, Yuan RY. Comparisons of forward and backward gait between poorer and better attention capabilities in early Parkinson's disease. Gait Posture. 2012;36(3):367–371.
25. Lin CC, Creath RA, Rogers MW. Variability of anticipatory postural adjustments during gait initiation in individuals with Parkinson disease. J Neurol Phys Ther. 2016;40(1):40–46.
26. Kakinuma S, Nogaki H, Pramanik B, Morimatsu M. Muscle weakness in Parkinson's disease: isokinetic study of the lower limbs. Eur Neurol. 1998;39(4):218–222.
27. Nogaki H, Kakinuma S, Morimatsu M. Muscle weakness in Parkinson's disease: a follow-up study. Parkinsonism Relat Disord. 2001;8(1):57–62.
28. Blin O, Ferrandez AM, Pailhous J, Serratrice G. Dopa-sensitive and dopa-resistant gait parameters in Parkinson's disease. J Neurol Sci. 1991;103(1):51–54.
29. Bohnen NI, Cham R. Postural control, gait, and dopamine functions in parkinsonian movement disorders. Clin Geriatr Med. 2006;22(4):797–812, vi.
30. Bryant MS, Rintala DH, Hou JG, et al. Gait variability in Parkinson's disease: influence of walking speed and dopaminergic treatment. Neurol Res. 2011;33(9):959–964.
31. McNeely ME, Duncan RP, Earhart GM. Medication improves balance and complex gait performance in Parkinson disease. Gait Posture. 2012;36(1):144–148.
32. Foreman KB, Addison O, Kim HS, Dibble LE. Testing balance and fall risk in persons with Parkinson disease, an argument for ecologically valid testing. Parkinsonism Relat Disord. 2011;17(3):166–171.
33. Curtze C, Nutt JG, Carlson-Kuhta P, Mancini M, Horak FB. Levodopa is a double-edged sword for balance and gait in people with Parkinson's disease. Mov Disord. 2015;30(10):1361–1370.
34. Horak FB, Nutt JG, Nashner LM. Postural inflexibility in parkinsonian subjects. J Neurol Sci. 1992;111(1):46–58.
35. Falvo MJ, Earhart GM. Six-minute walk distance in persons with Parkinson disease: a hierarchical regression model. Arch Phys Med Rehabil. 2009;90(6):1004–1008.
36. Canning CG, Ada L, Johnson JJ, McWhirter S. Walking capacity in mild to moderate Parkinson's disease. Arch Phys Med Rehabil. 2006;87(3):371–375.