ADVERTISEMENT

Issue Navigator

Volume 12 No. 11
Earn CME
Accepted Papers





Scientific Investigations

http://dx.doi.org/10.5664/jcsm.6270

Standing Balance and Spatiotemporal Aspects of Gait Are Impaired Upon Nocturnal Awakening in Healthy Late Middle-Aged and Older Adults

Amanda L. McBean, PhD1,2; Raymond P. Najjar, PhD1,2,3; Ronald A. Schuchard, PhD2,4; Courtney D. Hall, PhD5,6; Cheng-Ann Wang, BA2; Ban Ku, MS2; Jamie M. Zeitzer, PhD1,2
1Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford CA; 2Sierra Pacific Mental Illness Research, Education, and Clinical Center, VA Palo Alto Health Care System, Palo Alto, CA; 3Visual Neurosciences Group, Singapore Eye Research Institute, Singapore; 4Department of Neurosurgery, Stanford University; 5Auditory Vestibular Research Enhancement Award Program, James H. Quillen VA Medical Center, Mountain Home, TN; 6Department of Physical Therapy, East Tennessee State University, Johnson City, TN

ABSTRACT

Study Objectives:

Nocturnal awakenings may constitute a unique risk for falls among older adults. We describe differences in gait and balance between presleep and midsleep testing, and whether changes in the lighting environment during the midsleep testing further affect gait and balance.

Methods:

Twenty-one healthy, late middle-aged and older (64.7 ± 8.0 y) adults participated in this repeated-measures design consisting of four overnight laboratory stays. Each night, participants completed baseline visual acuity, gait, and balance testing. After a 2-h sleep opportunity, they were awakened for 13 min into one of four lighting conditions: very dim white light (< 0.5 lux); dim white light (∼28.0 lux); dim orange light (∼28.0 lux); and white room-level light (∼200 lux). During this awakening, participants completed the same sequence of testing as at baseline.

Results:

Low-contrast visual acuity significantly decreased with decreasing illuminance conditions (F(3,45) = 98.26, p < 0.001). Our a priori hypothesis was confirmed in that variation in stride velocity and center of pressure path length were significantly worse during the mid-sleep awakening compared to presleep baseline. Lighting conditions during the awakening, however, did not influence these parameters. In exploratory analyses, we found that over one-third of the tested gait and balance parameters were significantly worse at the midsleep awakening as compared to baseline (p < 0.05), and nearly one-quarter had medium to large effect sizes (Cohen d ≥ 0.5; r ≥ 0.3).

Conclusions:

Balance and gait are impaired during midsleep awakenings among healthy, late middle-aged and older adults. This impairment is not ameliorated by exposure to room lighting, when compared to dim lights.

Citation:

McBean AL, Najjar RP, Schuchard RA, Hall CD, Wang CA, Ku B, Zeitzer JM. Standing balance and spatiotemporal aspects of gait are impaired upon nocturnal awakening in healthy late middle-aged and older adults. J Clin Sleep Med 2016;12(11):1477–1486.


INTRODUCTION

Approximately one-third of adults older than 65 y experience at least one fall each year,1 and the 2-y prevalence of self-reported falls has increased over the past decade.2 The consequences of a fall for an older adult can be debilitating or even life-threatening. Falls accounted for 78% of injury admissions to hospitals in the United States among adults older than 65 y in 2009,3 are the leading cause of fatal injuries among adults older than 65 y,4 and result in premature institutionalization5 and substantial psychological morbidity including activity restriction6 and fear of falling.7 Falls are also costly; direct medical costs in adults older than 65 y totaled $200 million for fatal falls and $19 billion for nonfatal fall-related injuries in the United States.8

BRIEF SUMMARY

Current Knowledge/Study Rationale: Approximately one-third of adults older than 65 y experience at least one fall each year, and over one-fourth of these falls occur during the night. This study was conducted to determine whether midsleep arousals impair gait and standing balance, and whether changes in the lighting environment during the midsleep arousal could affect gait and standing balance.

Study Impact: This is to our knowledge the first study to report impairment of several aspects of both standing balance and gait during midsleep awakenings among healthy late middle-aged and older adults. This impairment is not ameliorated by exposure to room lighting, when compared to dim light.

Nocturia, or the frequent need to wake at night to void, is considered a risk factor for falls among older adults.9 A cohort study of 135,433 older adults reported that the presence of nocturia and urinary incontinence were the strongest predictors of falls, after older age, female sex, and a history of falls.10 These studies did not examine the time of day in which the fall occurred, so it is unknown whether nocturia increased daytime fall risk (potentially as a result of daytime sleepiness), nighttime fall risk, or both. However, an independent study reported that over one-fourth of falls among older adults occur during the night.11 In most individuals the majority of the nighttime hours are spent asleep with relatively low levels of activity compared to the daytime hours.12 Presumably, this decrease in activity also decreases fall opportunity overnight, making the fact that one-fourth of falls occur at night an even more substantial proportion. Of these nocturnal falls, over half were related to a visit to the toilet.11 Thus, it is likely that nocturnal awakenings, especially those with nocturia, pose a unique fall risk to the older adult.

The ability to maintain postural stability under different environmental conditions is critical to the ability to safely perform activities of daily living. Postural stability is a multisensory motor task that depends on reliable input from the vestibular, somatosensory, and visual systems. Deficits in one or more of these sensory systems, all of which undergo age-related changes, are likely to contribute to nocturnal falls in older individuals. Increased fall risk is linked to poor vision through impaired depth perception, contrast sensitivity, visual acuity, and visual fields.13 Dim lighting reduces the ability to acquire this visual information, and thus may impair postural and gait stability. It is well established that complete removal of visual input decreases postural stability14,15; however, the effects of graded light on postural stability in late middle-aged or older adults is not well known. One study demonstrated that reduced ambient light significantly impaired standing balance,14 whereas another study demonstrated reduced gait stability in low light.16

A low illuminance lighting environment, and its potential concomitant negative consequences on balance, is common upon arousal from sleep when arising to use the toilet. Additional sources of balance impairment might include incomplete arousal from sleep (i.e., sleep inertia), use of benzodiazepine hypnotic medications, and changes in blood pressure regulation. Two studies have assessed whether awakening from nocturnal sleep decreases standing balance and success on a tandem walk test among older adults. Neither study found an influence of awakening from sleep,17,18 but the effect of nocturnal awakening on spatial and temporal parameters of gait is unknown. No previous study has examined the effect of nocturnal awakening and different levels of environmental light conditions on gait and balance in healthy late middle-aged and older adults. Based on repeated associations with fall risk,19–22 the a priori primary outcome measures of interest for gait and standing balance were, respectively: (1) stride velocity variability at self-selected pace and (2) center of pressure (COP) displacement during standing with eyes open and eyes closed. We hypothesized that there would be greater variation in stride velocity and greater COP displacement during the midsleep awakening and in the dimly lit environments compared to baseline and room-level light, respectively. We also measured changes in visual acuity presleep and during the midsleep awakening, and across lighting conditions, with the hypothesis that vision would not change between the two time points, but would be progressively worse with decreasing illuminance. Finally, to direct future research on which balance and gait parameters were most affected by midsleep awakenings, we ran exploratory analyses on 29 balance and gait parameters in only the room-level white light condition to identify the parameters with the largest effect sizes.

METHODS

The protocol was approved by the Stanford University Institutional Review Board and all procedures adhered to the principles outlined in the Declaration of Helsinki. Written informed consent and Health Insurance Portability and Accountability Act authorization were obtained from participants prior to study participation. The study was posted on www.clinicaltrials.gov (#NCT01350505) on 05/06/11.

Participants

Twenty-one healthy, late middle-aged and older adults without sleep, balance, or psychiatric disorders (see Table 1 for demographics) were recruited through advertisements in local community newspapers, articles in senior newsletters, posters at local Veterans Affairs clinics, posters at local businesses, and word of mouth. Participants were included in the study based on successfully completing the first three lines of the 25% low-contrast visual acuity chart in very dim light (corrected-to-normal vision allowed) (Early Treatment Diabetic Retinopathy Study [ETDRS] charts [Precision Vision Inc., La Salle, IL]), normal 30° visual fields (Octopus 900, Haag-Streit, Mason, OH), stable health, no cognitive impairment (Mini Mental Examination Scores > 2323), nonsmoker, good sleeper (Pittsburgh Sleep Quality Index scores ≤ 524), no clinically significant symptoms of depression (Geriatric Depression Scale ≤ 525), normal color vision (Ishihara's Test for Colour Deficiency), normal hearing, no alcohol abuse (Alcohol Use Disorders Identification Test ≤ 1926) with proximal use tested upon admission to the laboratory (Alco-Screen 02, Chematics, Inc., North Webster, Indiana), and no illegal drug use with proximal use tested upon admission to the laboratory (DrugCheck3, Express Diagnostics, Blue Earth, MN). While 81% of study participants were regularly taking some type of medication (including vitamins and/or herbal medications), medication use was stable in the month prior to study enrollment and throughout the duration of their participation in the study.

Demographics and baseline information.

jcsm.12.11.1477.t01.jpg

table icon
Table 1

Demographics and baseline information.

(more ...)

Procedure

In this repeated-measures design, all participants were admitted into the laboratory for four separate overnight stays. Each laboratory stay differed only in the lighting environment into which the participant was awakened during the night (see next paragraphs). The four experimental lighting conditions were: very dim white light (< 0.4 ± 0.2 lux); dim white light (27.9 ± 0.6 lux); dim orange light (27.7 ± 0.6 lux); and room-level white light (202.4 ± 4.1 lux). The spectral characteristics of the orange and white light used are illustrated in Figure 1. Lighting was delivered using dimmable ceiling-mounted, high-power light-emitting diodes (LED) mounted with a UV filter (Rosco Cinegel #3114 Tough UV Filter), and a diffuser (Rosco Cinegel #3047, Light Velvet Frost) in addition to an orange filter (Rosco Roscolux #19, Fire) (Rosco, Sun Valley, CA) for the dim orange light. The order of the lighting condition was randomized across participants, except for the room-level white light, which occurred during participants' fourth and final visit, because this condition was added to the protocol after participants had completed the first three visits.

Spectral characteristics of the lighting environments.

jcsm.12.11.1477a.jpg

jcsm.12.11.1477a.jpg
Figure 1

Spectral characteristics of the lighting environments.

(more ...)

One week of at-home regular sleep-wake periods preceded each laboratory stay. Bed and wake times were selected upon consensus between the participant and research personnel and were at least 7 to 9 h apart so as to ensure participants were not significantly sleep deprived upon laboratory admission. Participants were required to go to bed and arise within 30 min of the agreed-upon bed and wake times. This schedule helps maximize the amplitude27 and stabilize the phase angle28 of the circadian system. Adherence was verified using self-reported sleep-wake logs and wrist actigraphy (Actiwatch 2, Philips Respironics, Murrysville, PA). The average sleep midpoint during the at-home week was calculated and used as the midpoint for an 8-h sleep opportunity in the laboratory.

Participants arrived at the Stanford University/VA Palo Alto Health Care System Sleep and Circadian Translational Research Laboratory 3 h before their scheduled bedtime. A research assistant prepared them for polysomnography according to the established 10-20 system including electroencephalography (C3/4, O1/2) referenced to an electrically neutral auricular electrode, bilateral electro-oculogram, submental electromyogram, and electrocardiogram (Siesta, Compumedics, Charlotte, NC). Recordings were analyzed according to the American Academy of Sleep Medicine's rules for sleep stage scoring29 using Prana Software (PhiTools, Chicago, IL). Participants remained in a time-isolation suite (no windows, no time cues, en suite bathroom, and regulated ambient temperature) for the duration of the stay until discharge the next morning. Lighting was maintained at normal room lighting (∼150 lux, in the direction of gaze; ∼250 lux maximal lighting) for the entire time they were in the time isolation suite, except during sleep periods (< 0.03 lux) and the experimental lighting conditions. Lighting was controlled by a technician from outside of the room.

Participants underwent a presleep baseline testing sequence consisting of visual acuity testing, standing balance testing, and gait testing. The lights were turned off and participants were instructed to go to bed at the precalculated habitual bedtime. Two hours into this sleep opportunity, participants were awakened by a technician into one of the four experimental lighting conditions and were kept awake for 13 min. During this awakening, participants were examined using the same test battery as occurred prior to sleep. After the 13 min awakening, the lights were turned off and the participants were permitted to return to sleep until their habitual wake time. Upon awakening in the morning, participants were given breakfast and then discharged.

Visual Acuity

During visual acuity testing, participants read a low-contrast ETDRS Chart from a distance of 3 m. The low-contrast ETDRS chart consists of 14 rows of 5 letters each. Going from the top to the bottom row, letters become progressively smaller and there is equal spacing (on a log scale) between letters and rows. Participants were instructed to read the eye chart from the top to the bottom; participants were stopped after they finished a row where they were unable to correctly identify at least three letters within the same row. The total number of correct letters was used as the measure of visual acuity. The contrast of the chart was adjusted (10% or 25%) depending on the participant's low contrast visual acuity ability to avoid a ceiling or floor effect.

Standing Balance and Gait

Balance and gait parameters were evaluated using a Zeno Walkway (ProtoKinetics, Havertown, PA). The Zeno Walkway is a 2 × 14 inch sensor pad with 16 levels of dynamic pressure detected through embedded 0.4-inch pressure sensors placed every 0.5 inches. The walkway is connected to a laptop computer by a serial interface cable in order to process and store data using ProtoKinetics Movement Analysis Software. Balance-related metrics and temporal and spatial measures of gait are calculated by the software (version 5.07C7). This system has been validated30 and previously used to assess gait in relationship to falls among older adults.31 The output parameters used in the current study and their descriptive statistics are listed in Tables 2 and 3. All but 6 of the 27 parameters have been previously associated with fall risk among older adults.20–22,32,33 We chose to include the six that have not been previously associated directly with fall risk (gait cycle time, % coefficient of variation (CoV) gait cycle time, stance time, %CoV single support %, COP X Range, COP Y Range) based on their close relationship with parameters that have been associated with falls: gait cycle time and the %CoV gait cycle time were chosen based on their similarity to mean and variation in stride velocity; mean stance time is of interest as the variation in stance time has been associated with falls22; similarly %CoV of single support % is of interest because the mean single support time has been associated with falls32; and standing COP X and Y Range were chosen because previous work has related the root mean square amplitude of the mediolateral and anterior-posterior acceleration to fall history.32

Definition of the Zeno Walkway output parameters used.

jcsm.12.11.1477.t02.jpg

table icon
Table 2

Definition of the Zeno Walkway output parameters used.

(more ...)

Descriptive statistics of the gait and balance parameters examined.

jcsm.12.11.1477.t03.jpg

table icon
Table 3

Descriptive statistics of the gait and balance parameters examined.

(more ...)

During the standing procedure, participants were asked to stand on the walkway in a comfortable foot position with their arms by their side and their eyes open for 30 sec and then with their eyes closed for 30 sec. Immediately after the standing procedure, participants walked along the walkway at a self-selected pace for three trials. One end of the walkway was positioned against a wall, such that participants performed a turn on the walkway at this end and the other end was open so that participants walked off the mat and continued walking for 2 m. To reduce artifact due to slowing down to turn, only the three walks that occurred from the wall to the open end of the walkway were analyzed. The first step of each walk was considered gait initiation and was excluded from analyses. These three walks were averaged together in the calculation of the gait parameters. ProtoKinetics Movement Analysis Software was used to derive all parameters.

Data Analysis

Data were analyzed using R version 3.1.334 and the R package lme4.35 A value of p < 0.05 was considered statistically significant. Data were checked for homoscedasticity and normality by visual inspections of residual plots and Shapiro-Wilk test. Differences in low-contrast visual acuity between pre-sleep and nocturnal awakenings in the room-level light condition were tested with a paired-samples t-test. Differences in low-contrast visual acuity between the four lighting conditions during the nocturnal awakenings were tested with a repeated-measures analysis of variance and post hoc paired t-tests with a Bonferroni correction for multiple comparisons (p = 0.008).

Given each participant contributed multiple data points, we used hierarchical multilevel modeling (HLM) to test the hypotheses that parameters of gait and balance differ upon awakening from nocturnal sleep compared to presleep baseline, and explore whether the change in values from presleep to midsleep differed by the lighting into which they were awakened. Using a simple linear regression is insufficient in this analyses because our data do not satisfy the assumption of independence of observations. The advantage of using HLM is that it accounts for shared variance within participants while modeling differences between participants.36 With 4 lighting conditions and 2 time points, the dataset featured 8 stacked cases for the 17 subjects that completed the protocol, and 6 stacked cases for the 4 subjects that did not return for the final condition (or a total of 160 data points). Specifically, crossed random-effects models, an extension of the classic HLM, were used to accommodate the nesting of the gait/balance parameter (Level 1) in the combination of participant and condition random effects (Level 2). A full model was fitted with fixed effects of time (i.e., are pre-sleep values different from values obtained after awakening, independent of lighting condition), condition (i.e., is the slope of the values from presleep to awakening different by the lighting condition into which they are awakened), and random effects of intercept (accounts for differences of presleep values among participants), and by-participant random slopes for time (accounts for differences in the change in values between presleep and awakening among participants). An unstructured covariate matrix was chosen to allow for correlation between the random effects. The initial model was overfitted, so the random slope term was removed from the model for subsequent analyses. The model tested for each dependent variable is:

  • Level 1 (item-level) model:

    • (Standing Balance/Gait Parameter)ij = β0ij + β1ij(Time) + εij

  • Level 2 (participant-level) model:

    • β0ij = γ00 + ζ0j

    • β1ij = γ10 + γ11(Condition)

Positive skewness was noted for all three dependent variables: coefficient of variation in stride velocity and center of pressure path length (both open and closed eyes). However, because modeling with and without the variables transformed (logarithmically) to closer approximate normality assumptions did not substantively change the results, untransformed values were used to facilitate interpretation. Modeling was achieved using likelihood ratio testing to compare full with stepwise reduced models.

A wide variety of parameters derivable from examination of gait and standing balance have been previously reported as proxies for predicting future falls in older adults. As such, we chose to look at an assortment of parameters we gathered from the Zeno Walkway in secondary exploratory analyses. These analyses were used to identify which parameters had the largest effect sizes in our study. Paired samples t-tests were used to determine whether balance and gait parameters differed between presleep and nocturnal awakenings in the room-level lighting condition. Wilcoxon signed-rank test was used when variables did not meet normality assumptions. Cohen d (small = 0.2, medium = 0.5, large = 0.8) and r (small = 0.1, medium = 0.3, large = 0.5) were used to interpret effect sizes.37,38

RESULTS

Four of the 21 participants did not return for the fourth and final condition, the room-level white light condition. The participants who did not return for the final condition did not differ in age, sex, race, or ethnicity from the participants that did return. During the 2-h sleep opportunity prior to the midsleep awakening, participants slept, on average, for 90.7 min (standard deviation ± 18.0); this total sleep time did not differ across lighting conditions.

Low-Contrast Visual Acuity

Notably, there was no difference between baseline low-contrast visual acuity obtained in the evening in normal room light and the low-contrast visual acuity measurements made upon awakening into the room-level light condition (t(16) = 1.38, p = 0.19). There was a significant difference across the four lighting conditions (F(3,45) = 98.26, p < 0.001) such that low-contrast visual acuity decreased with decreasing illuminance, as expected,39 but did not differ in equiluminant white and orange light (Figure 2).

Differences in visual acuity across lighting conditions during the nocturnal awakening.

Post hoc testing revealed graded difference by illuminance level.

jcsm.12.11.1477b.jpg

jcsm.12.11.1477b.jpg
Figure 2

Differences in visual acuity across lighting conditions during the nocturnal awakening.

(more ...)

Variation in Stride Velocity

A full model was fitted with the %CoV of stride velocity as the dependent variable with time (presleep vs. midsleep awakening) and condition (four lighting conditions) as independent variables. A reduced model that removed the effect of condition was fitted, but did not significantly differ from the full model (χ2(3) = 4.24, p = 0.24). As such, the condition term was removed and the model with just the fixed effect of time was compared to an intercepts-only model. This comparison was statistically significant (χ2(1) = 15.64, p < 0.001), supporting a significant effect of time on variation in stride velocity. The model with the fixed effect of time and random effect of participant intercept became the final model and is summarized in Table 4.

Results of the model of variation in stride velocity due to time and condition.

jcsm.12.11.1477.t04.jpg

table icon
Table 4

Results of the model of variation in stride velocity due to time and condition.

(more ...)

COP Path Length – Eyes Open

Similar to the model testing process for variation in stride velocity, a full model was fitted with COP path length during eyes open as the dependent variable and time (presleep vs. midsleep awakening) and condition (four lighting conditions) as independent variables. A reduced model with the condition (lighting) term removed did not significantly differ from the full model (χ2(3) = 5.78, p = 0.12) so the condition term was removed from further analyses. Comparison of the reduced model to an intercept only model revealed a significant difference (χ2(1) = 7.66, p = 0.006), supporting a significant effect of time on COP path length during the eyes open condition. The reduced model became the final model and is described in Table 5.

Results of the model of center of pressure path length (eyes open) due to time and condition.

jcsm.12.11.1477.t05.jpg

table icon
Table 5

Results of the model of center of pressure path length (eyes open) due to time and condition.

(more ...)

COP Path Length – Eyes Closed

Finally, the same procedure was followed for modeling the effects of time and lighting condition on COP path length during the eyes closed condition. The reduced model without the condition (lighting) term did not significantly differ from the full model (χ2(3) = 3.74, p = 0.29), supporting no effect of condition. A significant difference was found between the reduced model containing only the fixed effect of time and an intercept only model (χ2(1) = 5.24, p = 0.022), supporting an effect of time on COP path length in the eyes closed condition. The final model is described in Table 6.

Results of the model of center of pressure path length (eyes closed) due to time and condition.

jcsm.12.11.1477.t06.jpg

table icon
Table 6

Results of the model of center of pressure path length (eyes closed) due to time and condition.

(more ...)

Gait and Balance

While we had a priori hypotheses concerning the importance of variation in stride length (gait) and center of pressure path length (standing balance), we wanted to also examine the degree to which awakening an individual, independent of the lighting environment, would affect a variety of measures of gait and standing balance. Therefore, we examined many parameters reflecting these two behaviors and compared them between the presleep baseline in normal room light and the experimental awakening into identical lighting. Of the 27 measures of gait and balance, more than one-third (11) significantly differed upon a forced midsleep awakening and 8 had medium to large effect sizes (Figures 3 and 4).

Differences in normally distributed gait performance and standing balance measurements between presleep and nocturnal awakening in the room-level light condition expressed as Cohen d effect sizes.

*p < 0.05; EC, eyes closed; shading indicates values during the nocturnal awakening that were lower than presleep values; hatched columns indicate standing balance parameters.

jcsm.12.11.1477c.jpg

jcsm.12.11.1477c.jpg
Figure 3

Differences in normally distributed gait performance and standing balance measurements between presleep and nocturnal awakening in the room-level light condition expressed as Cohen d effect sizes.

(more ...)

Differences in nonnormally distributed gait performance and standing balance measurements between presleep and nocturnal awakening in the room-level light condition expressed as effect size r.

*p < 0.05; EO, eyes open; EC, eyes closed; all parameter medians were higher during the nocturnal awakening compared to presleep baseline; hatched columns indicate standing balance parameters.

jcsm.12.11.1477d.jpg

jcsm.12.11.1477d.jpg
Figure 4

Differences in nonnormally distributed gait performance and standing balance measurements between presleep and nocturnal awakening in the room-level light condition expressed as effect size r.

(more ...)

DISCUSSION

Our data support that awakening from a nocturnal sleep episode impairs aspects of both gait and standing balance among healthy late middle-aged and older adults. To the best of our knowledge, this is the first study to test these particular spatial and temporal aspects of gait performance during a forced nocturnal awakening in late middle-aged and older adults; however, our standing balance findings are in contradiction to those previously reported.17,18 These findings are contrary to two previous studies in which Frey et al.17 found balance was not impaired during a nocturnal awakening on the difficult tandem walking task and Zammit et al.18 reported no dynamic standing balance impairments as measured using a force plate during a nocturnal awakening. These studies were consistent with ours in participant health and age, and that the nocturnal awakenings all took place about 2 h after habitual sleep time; however, all studies used different equipment, methods, and measured different output parameters. Further, Frey et al.'s17 walking task was more difficult than our own, and it is possible that alertness circuits were sufficiently activated to more completely awaken their participants, minimizing differences from pre-sleep testing. Zammit et al.18 had fewer participants than our own study (n = 12), and all but one of their participants was female. These variations may account for the disparities in results. We believe, however, that our participant population and walking task is more ecologically valid with respect to the risk of falls during nocturnal awakenings among late middle-aged and older adults.

Results from our hierarchical level models indicate that variation in stride velocity and normative path length (in both eyes open and eyes closed conditions) were significantly worse during a nocturnal awakening compared to pre-sleep baseline, supporting our hypothesis that gait and balance are impaired upon nocturnal awakenings. Contrary to our hypothesis, however, the change in gait and balance values between time points did not differ across lighting condition. This finding contradicts that of previous research that dim lighting impaired postural stability14 and temporal and spatial measures of gait16 when compared to room-level lighting. Again, differences in protocols may provide an explanation for this disparity in results. Most notably, we tested the effects of light during a nocturnal awakening instead of during typical waking hours. It is possible that because balance and gait were significantly worse overnight compared to daytime baseline testing, the influence of light was mitigated in our study due to a floor effect.

Interestingly, despite no changes in gait performance or standing balance, low-contrast visual acuity decreased with decreasing illuminance. Despite the reduction in low-contrast visual acuity, there was no associated change in gait performance or standing balance and it may be that other aspects of vision (e.g., visual field, depth perception) were not significantly affected by the lighting and were utilized to maintain consistent gait performance and standing balance.40 It is also possible that, as our testing procedure involved walking on an even surface with no obstacles with evenly dispersed lighting throughout the room, low-contrast visual acuity was not the critical visual function to successfully perform the gait and standing balance tasks. As this study was the initial pilot study to determine important factors, the experimental task was kept to main exploratory factors that would help guide future studies. It would be an important area of future research to understand how gait performance might differ in different lighting in environments where there are more challenging visual and gait tasks with more complicated factors such as hazards, uneven surfaces, or stairs, as all of these represent obstacles an older adult might face when walking from the bedroom to the toilet at night.

Given the variety of parameters used as a proxy for falls, and in order to not exclude any potential parameters, for our secondary analyses we ran a large number of exploratory tests with data only from the room-level white light condition. Through these analyses, we identified six parameters that were both statistically significant at the p = 0.05 level and had medium to large effect sizes. Two of these were our a priori parameters of interest—variation in stride velocity and the standing balance COP path length (with eyes open). The remaining four parameters were variability in the gait parameters of double support time, single support time, swing percentage, and stance percentage. All four of these parameters are complementary to each other in that they are related to the time in which the feet were either in contact or not in contact with the floor during the walking protocol. Greater variability in each of these temporal parameters has been previously associated with either retrospectively or prospectively identified fallers among older adults.31,41–43 Our results support that intercycle variability in temporal parameters during walking assessments and deviation of the COP during standing assessments with eyes open are increased during nocturnal awakenings. Several mean walking performance assessments—stride velocity, stride width, stride length, and step length–also statistically worsened from the presleep to the nocturnal awakening assessments, but multiple comparisons without adjustment of p values combined with small effect sizes caution us from concluding a meaningful difference in these parameters within the present study. Results from these exploratory analyses, along with previous work,42 emphasize the importance of complementing mean assessments of gait with intercycle variability in parameters during walking assessments in future research, as it was the variability of specific parameters that had the largest effect sizes rather than the absolute average of the parameters.

We, however, only tested a single, approximate circadian phase and did not have a large enough sample size to discriminate the effect of awakening out of different sleep stages on measures of balance and gait. It is likely that balance and gait following a midsleep awakening will be affected through non-linear interactions between circadian phase, homeostatic pressure, and sleep stage. Our forced awakenings, therefore, cannot be generalized to spontaneous awakenings resulting from nocturia. The current study's sample size also precluded the analyses of effects of demographics such as age, sex, and race on balance and gait. Future research should aim for sufficient power to consider sleep stage upon awakening and other important demographics factors that might influence gait and balance. Additional measurements, such as orthostatic blood pressure, could be incorporated into future studies to help further understand the multifactorial nature of falls in older adults.

The current findings suggest that balance and gait are impaired in the middle of the night even among healthy late middle-aged and older adults, but that this impairment is not mitigated by exposure to room lighting. These results suggest that midsleep awakenings are in and of themselves a risk factor for falls in the healthy aging population. Future research is necessary to understand how midsleep awakenings affect balance and gait in a population of adults with poor balance, as well as in less healthy older adults, and whether specific types of nocturnal light exposure are useful countermeasures to such impairments in balance.

DISCLOSURE STATEMENT

This was not an industry supported study. Funding for this study was provided by the VA RR&D: 1I21RX000773-01A2, VA Sierra Pacific Mental Illness Research Education and Clinical Center. The authors have indicated no financial conflicts of interest.

ABBREVIATIONS

COP

center of pressure

CoV

coefficient of variation

EC

eyes closed

EO

eyes open

ETDRS

early treatment diabetic retinopathy study

HLM

hierarchical linear modeling

SD

standard deviation

ACKNOWLEDGMENTS

The authors thank Yvonne Quevedo for the recruitment of the participants.

REFERENCES

1 

Tinetti ME, Speechley M, Ginter SF, authors. Risk factors for falls among elderly persons living in the community. N Engl J Med. 1988;319:1701–7. [PubMed]

2 

Cigolle C, Ha J, Min L, et al., authors. The epidemiologic data on falls, 1998-2010: more older Americans report falling. JAMA Intern Med. 2015;175:443–5. [PubMed]

3 

Maxwell CA, Miller RS, Dietrich MS, Mion LC, Minnick A, authors. The aging of america: a comprehensive look at over 25,000 geriatric trauma admissions to United States hospitals. Am Surg. 2015;81:630–6. [PubMed]

4 

Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web-based Injury Statistics Query and Reporting System (WISQARS). 2005. Accessed November 4, 2015. Available at www.cdc.gov/injury/wisqars.

5 

Kiel DP, O'Sullivan P, Teno JM, Mor V, authors. Health care utilization and functional status in the aged following a fall. Med Care. 1991;29:221–8. [PubMed]

6 

Kosorok MR, Omenn GS, Diehr P, Koepsell TD, Patrick DL, authors. Restricted activity days among older adults. Am J Public Health. 1992;82:1263–7. [PubMed Central][PubMed]

7 

Zijlstra GAR, van Haastregt JCM, van Eijk JTM, van Rossum E, Stalenhoef PA, Kempen GIJM, authors. Prevalence and correlates of fear of falling, and associated avoidance of activity in the general population of community-living older people. Age Ageing. 2007;36:304–9. [PubMed]

8 

Stevens JA, Corso PS, Finkelstein EA, Miller TR, authors. The costs of fatal and non-fatal falls among older adults. Inj Prev. 2006;12:290–5. [PubMed Central][PubMed]

9 

Stewart RB, Moore MT, May FE, Marks RG, Hale WE, authors. Nocturia: a risk factor for falls in the elderly. J Am Geriatr Soc. 1992;40:1217–20. [PubMed]

10 

Rafiq M, McGovern A, Jones S, et al., authors. Falls in the elderly were predicted opportunistically using a decision tree and systematically using a database-driven screening tool. J Clin Epidemiol. 2014;67:877–86. [PubMed]

11 

Jensen J, Lundin-Olsson L, Nyberg L, Gustafson Y, authors. Falls among frail older people in residential care. Scand J Public Health. 2002;30:54–61. [PubMed]

12 

Benloucif S, Orbeta L, Ortiz R, et al., authors. Morning or evening activity improves neuropsychological performance and subjective sleep quality in older adults. Sleep. 2004;27:1542–51. [PubMed]

13 

Black A, Wood J, authors. Vision and falls. Clin Exp Optom. 2005;88:212–22. [PubMed]

14 

Brooke-Wavell K, Perrett LK, Howarth PA, Haslam RA, authors. Influence of the visual environment on the postural stability in healthy older women. Gerontology. 2002;48:293–7. [PubMed]

15 

Edwards A, author. Body sway and vision. J Exp Psychol. 1946;36:526–35. [PubMed]

16 

Figueiro MG, Plitnick B, Rea MS, Gras LZ, Rea MS, authors. Lighting and perceptual cues: effects on gait measures of older adults at high and low risk for falls. BMC Geriatr. 2011;11:49. [PubMed Central][PubMed]

17 

Frey DJ, Ortega JD, Wiseman C, Farley CT, Wright KP, authors. Influence of zolpidem and sleep inertia on balance and cognition during nighttime awakening: a randomized placebo-controlled trial. J Am Geriatr Soc. 2011;59:73–81. [PubMed]

18 

Zammit G, Wang-Weigand S, Peng X, authors. Use of computerized dynamic posturography to assess balance in older adults after nighttime awakenings using zolpidem as a reference. BMC Geriatr. 2008;8:15. [PubMed Central][PubMed]

19 

Piirtola M, Era P, authors. Force platform measurements as predictors of falls among older people - A review. Gerontology. 2006;52:1–16. [PubMed]

20 

Muir JW, Kiel DP, Hannan M, Magaziner J, Rubin CT, authors. Dynamic parameters of balance which correlate to elderly persons with a history of falls. PLoS One. 2013;8:e70566. [PubMed Central][PubMed]

21 

Maki BE, author. Gait changes in older adults: predictors of falls or indicators of fear? J Am Geriatr Soc. 1997;45:313–20. [PubMed]

22 

Mortaza N, Abu Osman NA, Mahdikhani M, authors. Are the spatio-temporal parameters of gait capable of distinguishing a faller from a non-faller elderly? Eur J Phys Rehabil Med. 2014;50:677–91. [PubMed]

23 

Folstein M, Folstein S, McHugh P, authors. “Mini-Mental State” a practice method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. [PubMed]

24 

Buysse D, Reynolds C III, Monk T, Berman S, Kupfer D, authors. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. J Psychiatr Res. 1989;28:193–213.

25 

Yesavage J, Brink T, Rose T, et al., authors. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. 1983;17:37–49.

26 

Saunders J, Aasland O, Babor T, de la Fuente J, Grant M, authors. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection on persons with harmful alcohol consumption. II. Addiction. 1993;88:791–804. [PubMed]

27 

Jewett ME, Kronauer RE, Czeisler CA, authors. Phase-amplitude resetting of the human circadian pacemaker via bright light: a further analysis. J Biol Rhythms. 1994;9:295–314. [PubMed]

28 

Duffy J, Dijk D, Hall E, Czeisler C, authors. Relationship of endogenous circadian melatonin and temperature rhythms to self-reported preference for morning or evening activity in young and older people. J Investig Med. 1999;47:141–50. [PubMed]

29 

Iber C, Ancoli-Israel S, Chesson AL Jr, Quan SF. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specification. 1st ed. Westchester, IL: American Academy of Sleep Medicine, 2007.

30 

Bilney B, Morris M, Webster K, authors. Concurrent related validity of the GAITRite walkway system for quantification of the spatial and temporal parameters of gait. Gait Posture. 2003;17:68–74. [PubMed]

31 

Callisaya ML, Blizzard L, Schmidt MD, et al., authors. Gait, gait variability and the risk of multiple incident falls in older people: a population-based study. Age Ageing. 2011;40:481–7. [PubMed]

32 

Doheny EP, Greene BR, Foran T, Cunningham C, Fan CW, Kenny RA, authors. Diurnal variations in the outcomes of instrumented gait and quiet standing balance assessments and their association with falls history. Physiol Meas. 2012;33:361–73. [PubMed]

33 

Menant JC, Schoene D, Sarofim M, Lord SR, authors. Single and dual task tests of gait speed are equivalent in the prediction of falls in older people: a systematic review and meta-analysis. Ageing Res Rev. 2014;16:83–104. [PubMed]

34 

R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2015. Accessed July 16, 2016. Available at http://www.r-project.org/.

35 

Bates D, Maechler M, Bolker B, Walker S, authors. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67:1–48.

36 

Woltman H, Feldstain A, Mackay JC, Rocchi M, authors. An introduction to hierarchical linear modeling. Tutor Quant Methods Psychol. 2012;8:52–69.

37 

Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd Edition. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.

38 

Field A. Discovering Statistics Using SPSS. Third Edition. London: SAGE Publications Ltd, 2009.

39 

Brabyn J, Schneck M, Haegerstrom-Portnoy G, Lott L, authors. The Smith-Kettlewell Institute (SKI) longitudinal study of vision function and its impact among the elderly: an overview. Optom Vis Sci. 2001;78:264–9. [PubMed]

40 

Turano K, Rubin GS, Herdman SJ, Chee E, Fried LP, authors. Visual stabilization of posture in the elderly: fallers vs. nonfallers. Optom Vis Sci. 1994;71:761–9. [PubMed]

41 

Hausdorff JM, Edelberg HK, Mitchell SL, Goldberger AL, Wei JY, authors. Increased gait unsteadiness in community-dwelling elderly failers. Arch Phys Med Rehabil. 1997;78:278–83. [PubMed]

42 

König N, Taylor WR, Armbrecht G, Dietzel R, Singh NB, authors. Identification of functional parameters for the classification of older female fallers and prediction of “first-time” fallers. J R Soc Interface. 2014;11:20140353. [PubMed Central][PubMed]

43 

Taylor ME, Delbaere K, Mikolaizak AS, Lord SR, Close JCT, authors. Gait parameter risk factors for falls under simple and dual task conditions in cognitively impaired older people. Gait Posture. 2013;37:126–30. [PubMed]

44 

National Council on Aging. 6 steps to protect your older loved one from a fall. Accessed June 5, 2016. Available at https://www.ncoa.org/healthy-aging/falls-prevention/preventing-falls-tips-for-older-adults-and-caregivers/6-steps-to-protect-your-older-loved-one-from-a-fall/.