Secondary Logo

Journal Logo

Research Articles

Predictive Factors of Fall-Related Activity Avoidance in People With Parkinson Disease—A Longitudinal Study With a 3-Year Follow-up

Nilsson, Maria H. PhD; Jonasson, Stina B. PhD; Zijlstra, G. A. Rixt PhD

Author Information
Journal of Neurologic Physical Therapy: July 2020 - Volume 44 - Issue 3 - p 188-194
doi: 10.1097/NPT.0000000000000316

Abstract

INTRODUCTION

Fall-related avoidance behavior is a sound strategy in hazardous situations, but unnecessary activity avoidance can result in disability and participation restrictions. In older people, in general, one longitudinal study identified predictive factors of fall-related activity avoidance by using the modified Survey of Activities and Fear of Falling in the Elderly (mSAFFE) and multivariable analyses. It showed that fear of falling predicted a change in activity avoidance after 6 months.1 Aging with a progressive disorder such as Parkinson disease (PD) induces specific challenges, not the least since balance problems are key features. Little is known about fall-related activity avoidance in people with PD, for example, prevalence over time and predictive factors.

In studies that included people with PD, 41% to 53% reported fall-related activity avoidance.2,3 Subanalyses showed that it is also reported by those without a history of falls (30%), although more common among single and recurrent fallers (50% and 57%, respectively).3 Cross-sectional PD studies showed that commonly avoided activities due to the risk of falling are “Go out when it is slippery,” “Go to a place with crowds,” and “Reaching for something above your head.”2–5 In people with PD, the severity of fall-related activity avoidance (ie, mSAFFE scores) is associated with a history of falls or near falls,3–5 fear of falling,3–5 unsteadiness while turning,4 and disease severity.3 Moreover, it is associated with physical functioning, mobility, gait speed, and the severity of parkinsonian motor symptoms.4 The literature shows conflicting results regarding sex.4,5 Only Rahman and coworkers5 performed multivariable analyses in their cross-sectional PD study; activities of daily living (ADL) and anxiety were significantly associated with fall-related activity avoidance, whereas depression and disease severity were not. Importantly, to the best of our knowledge, no longitudinal study has yet addressed fall-related activity avoidance in people with PD. Longitudinal studies are needed to identify predictive factors. Knowledge of predictive factors can nurture future intervention studies that address fall-related activity avoidance in people with PD.

This study examined how prevalence and severity of fall-related activity avoidance evolve over a 3-year period in people with PD. Specific aims were to identify predictive factors of fall-related activity avoidance (ie, mSAFFE scores) after 3 years, as well as to identify predictive factors of a change in fall-related activity avoidance by adjusting for baseline mSAFFE scores.

METHODS

This study is part of the larger project “Home and Health in People Ageing with PD” (HHPD), which includes baseline assessments (2013) and a 3-year follow-up (2016). Details of the project design (including sample size calculations) are published in a study protocol.6 The final sample size was in line with the power analysis.

Participants and Recruitment

Participants were recruited from 3 hospitals in Sweden. Prior publications presented flowcharts of the recruitment procedures, that is, for baseline7 and the 3-year follow-up.8

At baseline, 653 persons met the inclusion criterion of a PD diagnosis (ICD-10 [International Classification of Diseases, Tenth Revision]: G20.9) at least 1 year previously. Of those, 216 were not eligible due to difficulties in understanding or speaking Swedish (n = 10), severe cognitive difficulties (n = 91), residence outside Skåne County (n = 58), or other reasons that made them unable to give informed consent or take part in the majority of the data collection (eg, hallucinations, a recent stroke; n = 57). The remaining 437 persons were invited to participate. Of these, 22 were unreachable, 2 had a revised diagnosis, and 157 declined. One person was excluded because of extensive missing data, resulting in a sample of 255 participants at baseline.

All those who completed baseline assessments and had agreed to be contacted again (n = 255) were considered eligible for the 3-year follow-up. At that time, 22 participants were deceased, 3 had moved, and 1 ended up outside the follow-up window (3 years ± 3 months). Thus, 229 people were invited to participate. Of these, 8 were unreachable, 4 had a revised diagnosis, and 51 declined. One person was excluded because of extensive missing data and low data quality. The 3-year follow-up rendered a sample of 165 persons. At baseline, this sample (N = 165) was significantly younger and consisted of fewer women than those who were lost to follow-up (n = 90): on average, 68 versus 73 years old at baseline, P < 0.001 (independent-samples t test); 35% versus 49% women, P = 0.045 (Pearson's χ2 test). There was no significant difference regarding mean PD duration (independent-samples t test: 9 vs 10 years, P = 0.286), but those who were lost to follow-up had a more severe PD at baseline: median Hoehn & Yahr scale score during “on” state 3 versus 2, P = 0.006 (Mann-Whitney U test).

Specific inclusion criteria were applied to this study; only those with an mSAFFE total score at both time points were included (n = 153). Two additional persons were excluded since someone else had responded to mSAFFE, that is, not only assisting the participant in reading or writing. The final study sample size constituted 151 persons (36% women). At baseline, their mean (SD, min-max) age was 68 (8.8, 45-91) years and their mean PD duration was 9 (6.3, 2-43) years. See Table 1 for other details on the participants' characteristics at baseline.

Table 1. - Participants' Characteristics at Baseline (N = 151)
Variables Median (q1-q3)a Miss. Variables Median (q1-q3)1 Miss.
Fall-related activity avoidance (mSAFFE) 21 (18-28) Unsteadiness turning (yes), n (%) 75 (49.7)
Age, mean (SD), y 68 (8.8) Falls past 6 mo (yes), n (%) 64 (42.7) 1
Women, n (%) 55 (36.4) Near falls past 6 mo (yes), n (%) 76 (50.3)
ADL—needing help (PADLS, >2), n (%) 31 (20.5) Concerns about falling (FES-I) 23 (18-34) 3
Motor symptoms (UPDRS part III), mean (SD) 28 (12.0) 1 General self-efficacy (GSE), mean (SD) 30 (5.9) 1
Mobility device use indoors (yes), n (%) 30 (19.9) Cognitive functioning (MoCA), mean (SD) 26 (3.0) 2
Walking difficulties (Walk-12G) 12 (6-20.8) 3 Depressive symptoms (GDS-15) 2 (1-4) 5
Mobility (Timed Up & Go), s 11 (9.1-12.6) 4 Anxiety (NMSQ item 17, yes), n (%) 36 (24.0) 1
Dual-task balance problems (yes), n (%) 92 (60.9) Pain (yes), n (%) 96 (63.6)
Abbreviations: ADL, activities of daily living; FES-I, Falls Efficacy Scale–International (16-64, higher = worse); GDS-15, Geriatric Depression Scale (0-15, higher = worse); GSE, General Self-efficacy Scale (10-40, higher = better); Miss., n missing (- = no missings); MoCA, Montreal Cognitive Assessment (0-30, higher = better); mSAFFE, modified Survey of Activities and Fear of Falling in the Elderly (17-51, higher = worse); NMSQ, Nonmotor Symptoms Questionnaire; PADLS, Parkinson's Disease ADL Scale (those who scored >2 were classified as needing help from others in daily activities); UPDRS part III, Unified Parkinson's Disease Rating Scale part III (part III = motor examination, 0-108, higher = worse); Walk-12G, Generic Walk-12 (0-42, higher = worse).
aUnless otherwise stated.

The project was approved by the Regional Ethical Review Board in Lund, Sweden (nos. 2012/558; 2015/611). All participants gave their written informed consent.

Data Collection

Both data collection waves included a self-administered postal survey, which was administered about 10 days before the subsequent home visit that included interview-administered questions/questionnaires and clinical assessments. All data collectors underwent project-specific training.

In the larger project, similar data collections were carried out at both time points. The current study used data from baseline as well as the 3-year follow-up. Potential predictors (ie, independent variables) constituted only baseline values, whereas the dependent variable in the multivariable regression analyses was mSAFFE scores at the 3-year follow-up.

Fall-Related Activity Avoidance

The 17-item self-administered mSAFFE1 addresses activity avoidance due to the perceived risk of falling in relation to 17 activities (see Table 2). The instruction is as follows: “Please circle the opinion closest to your own to show whether there are any things you avoid doing in case you fall over. For each activity below, please circle an answer to show whether you never avoid the activity, whether you sometimes try to avoid doing it in case you fall over, or if you always avoid doing the activity in case you fall. Please answer to show whether you think you WOULD do the activity even if you currently don't need to do the activity (eg, if someone else does your shopping for you).” Response categories are as follows: 1 = never; 2 = sometimes; or 3 = always avoid. The total score can range from 17 to 51 (higher = worse). The mSAFFE is reliable and valid in people with PD.2,4

Table 2. - mSAFFE Scoring Distribution (Including Cronbach α and SEM) at Baseline and 3 Years Later (N = 151)a
Item Baseline, Mean (SD) 3-y Follow-up, Mean (SD)
1. Go to the shops 1.37 (0.584) 1.53 (0.719)
2. Clean your house 1.43 (0.617) 1.59 (0.768)
3. Prepare simple meals 1.21 (0.456) 1.39 (0.600)
4. Go to the doctor or dentist 1.15 (0.412) 1.26 (0.559)
5. Take a bath 1.42 (0.687) 1.62 (0.799)
6. Take a shower 1.20 (0.448) 1.32 (0.572)
7. Go for a walk 1.36 (0.558) 1.53 (0.710)
8. Go out when it is slippery 1.95 (0.760) 2.18 (0.767)
9. Visit a friend or relative 1.25 (0.461) 1.41 (0.614)
10. Go to a place with crowds 1.49 (0.642) 1.74 (0.755)
11. Go up and down stairs 1.42 (0.637) 1.64 (0.726)
12. Walk around indoors 1.17 (0.390) 1.24 (0.486)
13. Walk half a mile 1.58 (0.769) 1.80 (0.833)
14. Bend down to get something 1.42 (0.570) 1.57 (0.678)
15. Travel by public transport 1.47 (0.691) 1.58 (0.734)
16. Go out to a social event 1.36 (0.535) 1.55 (0.650)
17. Reach for something above your head 1.51 (0.672) 1.70 (0.738)
Total scoreb 23.8 (7.30) 26.6 (9.10)
Total min-max 17-50 17-51
Internal consistency (Cronbach α) 0.944 0.957
SEM 2.0 1.7
Abbreviations: mSAFFE, modified Survey of Activities and Fear of Falling in the Elderly (possible item score range 1-3, possible total score range 17-51, higher = worse); SEM, standard error of measurement.
a; .
bP < 0.001, paired-samples t test.

To investigate the proportion of participants who reported fall-related activity avoidance, a self-administered dichotomous (Yes/No) question was included: “Do you avoid activities due to a risk of falling?”

Potential Predictors (ie, Baseline Values)

The data collection included age (years) and sex (0 = man; 1 = woman). ADL were assessed using the single-item (self-administered) Parkinson's Disease ADL Scale (PADLS).9 The response categories range from 1 (no difficulties) to 5 (extreme difficulties), but each response option includes a more detailed description. PADLS scores were dichotomized10: “0 = not needing help from others in daily activities” (scores 1-2) versus “1 = needing help” (scores 3-5).

Motor symptoms related to PD were clinically assessed using the Unified PD Rating Scale (UPDRS part III, scored 0-108; higher = worse).11 Structured and interview-administered questions targeted the use of mobility devices indoors: canes; crutches; other walking devices without wheels (eg, walking frames); wheeled walkers (ie, rollators), and manual and powered wheelchairs, respectively. Those that reported using 1 or more mobility devices indoors were categorized as users.

The self-administered Generic Walk-12 (Walk-12G, scored 0-42; higher = worse) assesses perceived walking difficulties in everyday life.12 Mobility was assessed with the Timed Up & Go13; 3 trials (comfortable gait speed) were conducted. The mean time (seconds) of the final 2 trials was used.

An interview-administered dichotomous (Yes/No) question addressed the history of falls during the past 6 months. The European consensus definition of a fall was applied: “An event in which the respondent came to rest on the ground, floor, or lower level.”14 A self-administered question (Yes/No) concerned a history of near falls during the previous 6 months, defined as “a fall initiated but arrested by support from a wall, railing, or other person, etc.”15 Concerns about falling was assessed using the Falls Efficacy Scale–International (FES-I, scored 16-64; higher = worse).16 A dichotomous (Yes/No) self-administered question concerned dual tasking: “Do you experience balance problems while standing or walking when doing more than one thing at a time, eg, carrying a tray while walking?” Another question (Yes/No) concerned whether the participant perceived “unsteadiness while turning around while standing or while walking.”

General self-efficacy was assessed according to the self-administered General Self-efficacy Scale (GSE, scored 10-40; higher = better).17 Cognitive functioning was clinically assessed using the Montreal Cognitive Assessment (MoCA, scored 0-30; higher = better).18 Depressive symptoms were assessed with the interview-administered 15-item Geriatric Depression Scale (GDS-15, scored 0-15; higher = worse).19 Anxiety was assessed with a dichotomous (Yes/No) item (no. 17) of the self-administered Nonmotor Symptoms Questionnaire (NMSQuest).20 Pain was assessed by the interview-administered dichotomous (Yes/No) question: “Are you bothered by pain?”

Statistical Analyses

Baseline characteristics were analyzed using descriptive statistics (means and SD, medians and q1-q3, or percentages, depending on data level) for all measures. Next, baseline and 3-year follow-up data on fall-related activity avoidance were compared. McNemar's exact test was used for the dichotomous question on fall-related activity avoidance and Wilcoxon matched-pairs signed ranks test for mSAFFE scores. Exact, 2-tailed P values are reported. For mSAFFE, we also report Cronbach α and standard error of measurement (SEM) in Table 2.

Regression Analyses

Simple (univariable) linear regression analyses were conducted to investigate the unadjusted relationship of each independent variable with mSAFFE scores at the 3-year follow-up.

Pearson (r) and Spearman (rs) correlations were used to assess relationships among independent variables in order to detect any multicollinearity (r or rs >0.7). There was multicollinearity (r = 0.85) between FES-I and Walk-12G. Therefore, 2 separate multivariable models were applied including all independent variables with a P value less than 0.3 in the univariable linear regression analyses. Model 1 included the FES-I but not Walk-12G, whereas model 2 included the Walk-12G but not FES-I. All independent variables were entered simultaneously. Probability values (P values) for all variables were inspected; the variable with the highest P value was manually removed. This procedure continued until all independent variables in the model had P values less than 0.1, which became the final model. The strength of the relationship between each independent variable and the dependent variable was assessed by the standardized regression coefficient (β). Residuals were analyzed graphically as a validation of the final multivariable models. Moreover, for each of the 2 models, we reran the analyses and controlled for baseline mSAFFE scores. These results reflect predictive factors of a change in mSAFFE scores over a 3-year period. That is, these models identify predictive factors given the mSAFFE score at baseline.

All statistical analyses were performed using SPSS v. 24 software for Windows (IBM Corporation, Armonk, New York).

RESULTS

In the total sample (N = 151), the proportion of participants who reported fall-related activity avoidance increased from 34% to 50% 3 years later (P < 0.001). The median mSAFFE score increased from 21 (q1-q3, 18-28) to 25 (18-33) 3 years later (P < 0.001). This difference exceeds the measurement error (ie, SEM) of 2 points (see Table 2).

At baseline, the top 5 activities of the mSAFFE that were most commonly avoided were as follows: (1) “Go out when it is slippery” (mean score 1.95); (2) “Walk half a mile” (ie, 1 km in the Swedish version) (mean score 1.58); (3) “Reach for something above your head” (mean score 1.51); (4) “Go to a place with crowds” (mean score 1.49); and (5) “Travel by public transport” (mean score 1.47). Three years later, the top 5 activities were as follows: (1) “Go out when it is slippery” (mean score 2.18); (2) “Walk half a mile” (mean score 1.80); (3) “Go to a place with crowds” (mean score 1.74); (4) “Reach for something above your head” (mean score 1.70); and (5) “Go up and down stairs” (mean score 1.64). Details are provided in Table 2, which lists mSAFFE item mean scores at baseline and at the 3-year follow-up.

Simple Regression Analyses

Simple linear regression analyses are presented in Table 3. All potential predictors were significantly (P < 0.05) associated with fall-related activity avoidance at the 3-year follow-up.

Table 3. - Simple Linear Regression Analyses With mSAFFE Scores at the 3-Year Follow-up as the Dependent Variable (N = 151)a
Independent Variables B 95% CI P Independent Variables B 95% CI P
Age, y 0.370 0.213 to 0.528 <0.001 Falls past 6 mo (yes = 1) 3.52 0.590 to 6.44 0.019
Sex (women = 1) 3.25 0.246 to 6.26 0.034 Near falls past 6 mo (yes = 1) 4.64 1.80 to 7.48 0.002
ADL—needing help (PADLS, >2 = 1) 9.18 5.86 to 12.5 <0.001 Concerns about falling (FES-I) 0.597 0.513 to 0.682 <0.001
Motor symptoms (UPDRS part III) 0.225 0.107 to 0.344 <0.001 General self-efficacy (GSE) −0.534 −0.767 to −0.301 <0.001
Mobility device use indoors (yes = 1) 7.06 3.56 to 10.6 <0.001 Cognitive functioning (MoCA) −0.673 −1.15 to −0.190 0.007
Walking difficulties (Walk-12G) 0.543 0.432 to 0.654 <0.001 Depressive symptoms (GDS-15) 1.28 0.714 to 1.85 <0.001
Mobility (Timed Up & Go), s 0.773 0.531 to 1.02 <0.001 Anxiety (NMSQ item 17, yes = 1) 5.06 1.71 to 8.41 0.003
Dual-task balance problems (yes = 1) 8.70 6.04 to 11.4 <0.001 Pain (yes = 1) 6.24 3.36 to 9.12 <0.001
Unsteadiness turning (yes = 1) 9.96 7.50 to 12.4 <0.001
Abbreviations: ADL, activities of daily living; FES-I, Falls Efficacy Scale–International (16-64, higher = worse); GDS-15, Geriatric Depression Scale (0-15, higher = worse); GSE, General Self-efficacy Scale (10-40, higher = better); MoCA, Montreal Cognitive Assessment (0-30, higher = better); mSAFFE, modified Survey of Activities and Fear of Falling in the Elderly (17-51, higher = worse); NMSQ, Nonmotor Symptoms Questionnaire; PADLS, Parkinson's Disease ADL Scale (those who scored >2 were classified as needing help from others in daily activities); UPDRS part III, Unified Parkinson's Disease Rating Scale part III (part III = motor examination, 0-108, higher = worse); Walk-12G, Generic Walk-12 (0-42, higher = worse).
aN = 146-151, depending on internal missing data (see Table 1).

Multivariable Regression Analyses

Model 1 (FES-I Included, Without Walk-12G)

The final model of model 1 resulted in 4 significant independent variables explaining 63% of the variance in mSAFFE scores at the 3-year follow-up (see Table 4). The strongest predictive factor (as assessed by β) was concerns about falling (β = 0.589), followed by pain (β = 0.161), unsteadiness while turning (β = 0.137), and age (β = 0.136). When adjusting for baseline mSAFFE scores, the former 4 independent variables remained significant and explained 64% of the variance, but the order changed. Concerns about falling remained the strongest predictor (β = 0.375), followed by age (β = 0.154), pain (β = 0.148), and unsteadiness while turning (β = 0.125).

Table 4. - Multivariable Linear Regression Analyses With mSAFFE Scores at the 3-Year Follow-up as the Dependent Variable: Model 1a
Independent Variablesb Unadjusted for mSAFFE Scores at Baseline (n = 148) Adjusted for mSAFFE Scores at Baseline (n = 148)
B 95% CI β P B 95% CI β P
Concerns about falling (FES-I) 0.465 0.362 to 0.567 0.589 <0.001 0.296 0.115 to 0.477 0.375 0.002
Pain (yes = 1) 3.03 1.00 to 5.07 0.161 0.004 2.79 0.775 to 4.81 0.148 0.007
Unsteadiness turning (yes = 1) 2.49 0.207 to 4.77 0.137 0.033 2.27 0.009 to 4.52 0.125 0.049
Age, y 0.141 0.025 to 0.256 0.136 0.017 0.159 0.044 to 0.274 0.154 0.007
R 2/adjusted R 2 62.6%/61.5% 63.8%/62.5%
Abbreviations: FES-I, Falls Efficacy Scale–International (16-64, higher = worse); mSAFFE, modified Survey of Activities and Fear of Falling in the Elderly (17-51, higher = worse).
aModel 1 included the FES-I but not Walk-12G (Generic Walk-12).
bThe following 16 independent variables were included: age; sex; need help in activities of daily living (PADLS); motor symptoms (UPDRS part III); mobility device use indoors; mobility (Timed Up & Go); dual-task balance problems; unsteady turning; falls past 6 months; near falls past 6 months; concerns about falling (FES-I); general self-efficacy (GSE); cognitive functioning (MoCA); depressive symptoms (GDS-15); anxiety (NMSQ item 17); and pain.

Model 2 (Walk-12G Included, Without FES-I)

The final model of model 2 included 4 significant independent variables, which explained 50% of the variance in mSAFFE scores at the 3-year follow-up (see Table 5). The strongest predictive factor was perceived walking difficulties (β = 0.392), followed by age (β = 0.238), unsteadiness while turning (β = 0.198), and pain (β = 0.184). When adjusting for baseline mSAFFE scores, 3 significant independent variables explained 60% of the variance in a change of mSAFFE scores. The strongest predictive factor was age (β = 0.193), followed by unsteadiness while turning (β = 0.165) and pain (β = 0.152).

Table 5. - Multivariable Linear Regression Analyses With mSAFFE Scores at the 3-Year Follow-up as the Dependent Variable: Model 2a
Independent Variablesb Unadjusted for mSAFFE Scores at Baseline (n = 148) Adjusted for mSAFFE Scores at (n = 151)
B 95% CI β P B 95% CI β P
Walking difficulties (Walk-12G) 0.341 0.209 to 0.473 0.392 <0.001
Age, y 0.247 0.117 to 0.377 0.238 <0.001 0.200 0.085 to 0.316 0.193 0.001
Unsteadiness turning (yes = 1) 3.58 0.861 to 6.30 0.198 0.010 2.99 0.699 to 5.29 0.165 0.011
Pain (yes = 1) 3.46 1.10 to 5.83 0.184 0.004 2.86 0.771 to 4.96 0.152 0.008
R 2/adjusted R 2 50.0%/48.6% 60.1%/59.0%
Abbreviations: mSAFFE, modified Survey of Activities and Fear of Falling in the Elderly (17-51, higher = worse); Walk-12G, Generic Walk-12 (0-42, higher = worse).
aModel 2 included the Walk-12G but not the FES-I (Falls Efficacy Scale-International).
bThe following 16 independent variables were included: age; sex; need help in activities of daily living (PADLS); motor symptoms (UPDRS part III); mobility device use indoors; walking difficulties (Walk-12G); mobility (Timed Up & Go); dual-task balance problems; unsteady turning; falls past 6 months; near falls past 6 months; general self-efficacy (GSE); cognitive functioning (MoCA); depressive symptoms (GDS-15); anxiety (NMSQ item 17); and pain.

DISCUSSION

This longitudinal study showed that fall-related activity avoidance increases over a time period of 3 years in people with PD. Compared with a multivariable regression model that included perceived walking difficulties, a model that included concerns about falling explained a somewhat larger variability in mSAFFE scores. The latter model showed that concerns about falling, pain, unsteadiness while turning, and age predicted fall-related activity avoidance, including a change in fall-related activity avoidance over 3 years' time.

Both the prevalence and severity of fall-related activity avoidance increased over time. This suggests that it might be important to address fall-related activity avoidance at clinical follow-ups. Although activity avoidance can be a sound strategy in hazardous circumstances, it might result in sedentary behavior and induce social isolation. In fact, some of the most commonly avoided activities were “Walk half a mile” and “Go to a place with crowds.”

Several modifiable factors independently predicted fall-related activity avoidance in people with PD. In the first multivariable model, concerns about falling were the strongest predictor. This also applied for a change in fall-related activity avoidance. Although these are novel findings within the PD field, they are in line with a study in community-living older people.1 The present results suggest that it is important to address concerns about falling among people with PD, not least since other PD studies have shown that fear of falling is a risk factor for future falls21 and a barrier to physical activity,22 exercise,23 and perceived participation.24 There is no consensus on how to address fear of falling in people with PD. A Cochrane review reported that exercise interventions are associated with a small and short-term reduction in fear of falling in older people.25 It has been suggested that combining psychotherapy and exercise might be the way ahead,26 given the effects of cognitive behavioral interventions.27,28 This requires further study.

In this study, pain predicted fall-related activity avoidance, including a change over time. This is in line with a cross-sectional study of older people.29 In people with PD, pain has been shown to predict future falls21 and interfere with activities, including walking.30 Sixty-four percent of our participants reported that they were bothered by pain. This corresponds with a review article that reported a pain prevalence of 60% in people with PD.31 Pain is thus a common symptom in PD, and by addressing pain, adverse outcomes, such as fall-related activity avoidance, might be prevented.

In both multivariable models, unsteadiness while turning predicted fall-related activity avoidance in people with PD. This is not surprising, since balance problems are a key feature of PD. Postural control during turning is commonly affected32; it can be altered already in the early stages of PD,33 and turning is associated with compromised stability32 and experiences of fear of falling.34 All things considered, this implies that turning deficits should be addressed early on and turns should be incorporated in balance training for people with PD. By doing so, activity avoidance may potentially be reduced.

In our second multivariable model, walking difficulties was the strongest predictor of fall-related activity avoidance 3 years later. However, when adjusting for baseline values (ie, mSAFFE scores), walking difficulties did not predict a change in fall-related activity avoidance. The latter was probably caused by multicollinearity between Walk-12G scores and mSAFFE scores at baseline. All things considered, this suggests that walking difficulties should be addressed if the aim is to work proactively against fall-related activity avoidance. In people with PD, walking difficulties and balance problems are important referral criteria to physiotherapy.35

Several variables that were significantly associated with fall-related activity avoidance in univariable analyses did not turn out as significant predictors when using multivariable analyses. This can be interpreted as illustrating a major drawback when relying solely on univariable regression analyses or bivariate analyses, as the results are often changed when the complexity of several variables are considered simultaneously (as in real life). Yet, some of these variables can still be important for the individual with PD.

Strengths of this study include the longitudinal design and the use of multivariable regression analyses. Moreover, the sample size allowed us to study the predictive value of a broad variety of factors. Still, there may be additional factors that influence fall-related activity avoidance in people with PD, such as fatigue and lack of insight into one's abilities.

Limitations

In longitudinal studies of people with a progressive disorder, dropouts may hamper the external validity of the findings. The current study showed that those included in the 3-year follow-up were younger and had less severe PD than the dropouts, but there was no significant difference regarding PD duration.

The potential predictors were assessed at baseline; this implies that changes over time in these independent variables were not taken into account when studying predictive factors of fall-related activity avoidance over a 3-year period. This was however not part of the scope of the present study.

Although the identified predictors point in a direction of what to target to prevent unnecessary fall-related activity avoidance, the benefit of doing so needs to be shown in future intervention studies. The knowledge gained by this study may facilitate developing intervention studies that address fall-related activity avoidance in people with PD.

CONCLUSIONS

Both the prevalence and severity of fall-related activity avoidance increase in people with PD after 3 years. This study suggests that the following modifiable factors should be addressed if the aim is to target fall-related activity avoidance in people with PD: concerns about falling, pain, unsteadiness while turning, and walking difficulties. Future intervention studies are needed to support or refute the present findings.

ACKNOWLEDGMENTS

The authors thank Susann Ullén, PhD, Clinical Studies Sweden—Forum South, Skåne University Hospital, Lund, Sweden, for valuable statistical advice.

REFERENCES

1. Yardley L, Smith H. A prospective study of the relationship between feared consequences of falling and avoidance of activity in community-living older people. Gerontologist. 2002;42(1):17–23.
2. Jonasson SB, Nilsson MH, Lexell J. Psychometric properties of four fear of falling rating scales in people with Parkinson's disease. BMC Geriatr. 2014;14:66.
3. Kader M, Iwarsson S, Odin P, Nilsson MH. Fall-related activity avoidance in relation to a history of falls or near falls, fear of falling and disease severity in people with Parkinson's disease. BMC Neurol. 2016;16:84.
4. Nilsson MH, Drake AM, Hagell P. Assessment of fall-related self-efficacy and activity avoidance in people with Parkinson's disease. BMC Geriatr. 2010;10(1):78.
5. Rahman S, Griffin HJ, Quinn NP, Jahanshahi M. On the nature of fear of falling in Parkinson's disease. Behav Neurol. 2011;24(3):219–228.
6. Nilsson MH, Iwarsson S. Home and health in people ageing with Parkinson's disease: study protocol for a prospective longitudinal cohort survey study. BMC Neurol. 2013;13:142.
7. Kader M, Ullen S, Iwarsson S, Odin P, Nilsson MH. Factors contributing to perceived walking difficulties in people with Parkinson's disease. J Parkinsons Dis. 2017;7(2):397–407.
8. Kader M, Jonasson SB, Iwarsson S, Odin P, Nilsson MH. Mobility device use in people with Parkinson's disease: a 3-year follow-up study. Acta Neurol Scand. 2018;138(1):70–77.
9. Hobson JP, Edwards NI, Meara RJ. The Parkinson's disease activities of daily living scale: a new simple and brief subjective measure of disability in Parkinson's disease. Clin Rehabil. 2001;15(3):241–246.
10. Lindholm B, Hagell P, Hansson O, Nilsson MH. Factors associated with fear of falling in people with Parkinson's disease. BMC Neurol. 2014;14(1):19.
11. Fahn S, Elton R; Members of the UPDRS Development Committee. Unified Parkinson's Disease Rating Scale. In: Fahn S, Marsden CD, Calne D, Goldstein M, eds. Recent Developments in Parkinson's Disease. Vol 2. Florham Park, NJ: MacMillan Healthcare Information; 1987:153–163, 293-304.
12. Bladh S, Nilsson MH, Hariz GM, Westergren A, Hobart J, Hagell P. Psychometric performance of a Generic Walking Scale (Walk-12G) in multiple sclerosis and Parkinson's disease. J Neurol. 2012;259(4):729–738.
13. 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.
14. Lamb SE, Jorstad-Stein EC, Hauer K, Becker C. Development of a common outcome data set for fall injury prevention trials: the Prevention of Falls Network Europe consensus. J Am Geriatr Soc. 2005;53(9):1618–1622.
15. Gray P, Hildebrand K. Fall risk factors in Parkinson's disease. J Neurosci Nurs. 2000;32(4):222–228.
16. Yardley L, Beyer N, Hauer K, Kempen G, Piot-Ziegler C, Todd C. Development and initial validation of the Falls Efficacy Scale-International (FES-I). Age Ageing. 2005;34(6):614–619.
17. Schwarzer R, Jerusalem M. Generalized Self-efficacy Scale. In: Wein-man J, Wright S, Johnston M, eds. Measures in Health Psychology: A User's Portfolio. Causal and Control Beliefs. Windsor, England: NFER-NELSON; 1995:35–37.
18. Nasreddine Z, Phillips N, Bedirian V, Charbonneau S, Whitehead V, Collin I. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695–699.
19. Sheikh J, Yesavage J. Geriatric Depression Scale (GDS): recent evidence and development of a shorter version In: Brink TL, ed. Clinical Gerontology: A Guide to Assessment and Intervention. New York, NY: Haworth Press; 1986:165–173.
20. Chaudhuri KR, Martinez-Martin P, Schapira AH, et al. International multicenter pilot study of the first comprehensive self-completed Nonmotor Symptoms Questionnaire for Parkinson's disease: the NMSQuest study. Mov Disord. 2006;21(7):916–923.
21. Lindholm B, Hagell P, Hansson O, Nilsson MH. Prediction of falls and/or near falls in people with mild Parkinson's disease. PLoS One. 2015;10(1):e0117018.
22. Bryant MS, Rintala DH, Hou JG, Protas EJ. Relationship of falls and fear of falling to activity limitations and physical inactivity in Parkinson's disease. J Aging Phys Act. 2015;23(2):187–193.
23. Ellis T, Boudreau JK, Deangelis TR, et al. Barriers to exercise in people with Parkinson disease. Phys Ther. 2013;93(5):628–636.
24. Thordardottir B, Nilsson MH, Iwarsson S, Haak M. “You plan, but you never know”—participation among people with different levels of severity of Parkinson's disease. Disabil Rehabil. 2014;36(26):2216–2224.
25. Kumar A, Delbaere K, Zijlstra GA, et al. Exercise for reducing fear of falling in older people living in the community: Cochrane systematic review and meta-analysis. Age Ageing. 2016;45(3):345–352.
26. Whipple MO, Hamel AV, Talley KMC. Fear of falling among community-dwelling older adults: a scoping review to identify effective evidence-based interventions. Geriatr Nurs. 2018;39(2):170–177.
27. Zijlstra GA, van Haastregt JC, Ambergen T, et al. Effects of a multicomponent cognitive behavioral group intervention on fear of falling and activity avoidance in community-dwelling older adults: results of a randomized controlled trial. J Am Geriatr Soc. 2009;57(11):2020–2028.
28. Dorresteijn TA, Zijlstra GA, Ambergen AW, Delbaere K, Vlaeyen JW, Kempen GI. Effectiveness of a home-based cognitive behavioral program to manage concerns about falls in community-dwelling, frail older people: results of a randomized controlled trial. BMC Geriatr. 2016;16:2.
29. Stubbs B, Eggermont LH, Patchay S, Schofield PA. Pain interference is associated with psychological concerns related to falls in community-dwelling older adults: multisite observational study. Phys Ther. 2014;94(10):1410–1420.
30. Rana AQ, Qureshi ARM, Haris A, et al. Negative impact of severity of pain on mood, social life and general activity in Parkinson's disease. Neurol Res. 2018;40(12):1054–1059.
31. Rana AQ, Kabir A, Jesudasan M, Siddiqui I, Khondker S. Pain in Parkinson's disease: analysis and literature review. Clin Neurol Neurosurg. 2013;115(11):2313–2317.
32. Conradsson D, Paquette C, Franzen E. Turning stability in individuals with Parkinson disease. J Neurol Phys Ther. 2018;42(4):241–247.
33. Song J, Sigward S, Fisher B, Salem GJ. Altered dynamic postural control during step turning in persons with early-stage Parkinson's disease. Parkinsons Dis. 2012;2012:386962.
34. Jonasson SB, Nilsson MH, Lexell J, Carlsson G. Experiences of fear of falling in persons with Parkinson's disease—a qualitative study. BMC Geriatr. 2018;18(1):44.
35. Domingos J, Keus SHJ, Dean J, de Vries NM, Ferreira JJ, Bloem BR. The European physiotherapy guideline for Parkinson's disease: implications for neurologists. J Parkinsons Dis. 2018;8(4):499–502.
Keywords:

accidental falls; fear of falling; pain; postural balance; walking

Supplemental Digital Content

© 2020 Academy of Neurologic Physical Therapy, APTA.