What is the relation between fear of falling and physical activity in older adults?

Abstract

BACKGROUND

Fear of falling (FOF) relates to participation in exercise, and FOF is related to reported physical function, but the association between FOF and total amount of daily activity performed by older adults has not been examined.

OBJECTIVE

The primary objective if this study is to describe the associations between FOF and total daily activity in older adults.

DESIGN

Cross-sectional observational study.

SETTING

Community-dwelling older adults.

PARTICIPANTS

Older adults over 64 who were independent in ambulation with or without and assistive device.

MEASUREMENTS

Fear of falling was defined by self-reported fear ratings using the Survey of Activities and Fear of Falling in the Elderly (SAFFE) and self-reported fear status determined by response to the question “Are you afraid of falling?” Physical function was assessed using the Late Life Function and Disability Instrument (LLFDI). Physical Activity was recorded using an accelerometer worn on the waist for seven consecutive days and mean daily counts of activity per minute were averaged over the seven-day period.

RESULTS

Fear ratings were related to total daily activity, r=−.26, p=.02. The relation was not as strong as the relation of function and physical activity, r=.45, p<.001. When stratified by exercise status or when stratified by functional status, fear was no longer related to total daily activity. Physical function explained 19% of the variance in physical activity, while the addition of fear status did not add to the explained variance in physical activity.

CONCLUSION

Fear of falling is related to total daily physical activity, but FOF was not independently associated with physical activity when accounting for physical function. Some fear of falling may be reported as a limitation in function.

Fear of falling (FOF) is known to be a modifier of physical performance in older adults. It is associated with a decrease in self-reported and performance-based physical function (PF).¹ Fear of falling is also associated with less participation in outdoor physical activity such as walking programs and sporting activities.^{2, 3} In these studies, walking programs, formal exercise and outdoor sports represented physical activity (PA). However, in older adults these activities account for only a small portion of usual daily physical activity.⁴ Among older adults a significant amount of the physical activity performed is low intensity activity, unlike the outdoor and sporting activities examined in previous reports.^{3, 4}

While PA and exercise share common elements, including bodily movement and energy expenditure, it is important to note that exercise is only one subset of PA.⁵ Therefore, total daily PA may include exercise as well as functional tasks like rising from bed, walking from room to room, driving a car, and shopping, the latter activities common among older adults. For many older adults, daily PA does not include exercise and in fact may be restricted to the activity involved in daily physical functions. The relationship of daily PA and physical function have been described based on self-reported daily PA but the relationship is not known for a performance-based measure of PA (eg. Accelerometry) and self-reported function.⁶ Accelerometry-based measures of PA provides the means to describe ‘what activity the person actually does’ and not the ‘the PA the person perceives they do’ or estimated from self-reported types of activities. While prior studies have examined the relations among FOF and exercise, and FOF and physical function,^{2, 3, 7, 8} the association between FOF and the total amount of daily activity performed by older adults has not been described. The primary aim of this study was to describe the associations among FOF and total daily activity in adults older than 65 years. To be able to explore relationships between exercise, physical function and PA in older adults in the present study to the established relationships, we stratified the sample based on exercise and based on physical function and examined the relationship of FOF and PA within the stratified groups.

While community-dwelling older adults have generally demonstrated low amounts of daily PA based on the usual record of outdoor activities, exercise and sports, it is possible that some older adults acquire substantial daily PA in the repeated performance of usual daily physical functions. By using a performance-based measure to record total daily PA, we may capture the PA of daily functions. If usual daily physical function is a primary form of total daily PA, making activity recommendations for health promotion that incorporate physical functions of daily living may be a means to enhance the potential for older adults to meet recommended daily PA guidelines.

METHODS

Community-dwelling older adults were recruited from the registry of the University of Pittsburgh Claude D. Pepper Older Adults Independence Center (OAIC) to participate in a longitudinal study describing mobility. Efforts have been made to populate the registry with older adults representative of the local metropolitan area. This study was approved by the Institutional Review Board at the University of Pittsburgh and all subjects provided informed consent to participate.

Participants

Subjects were included in this observational, cross-sectional analysis if they were 64 years of age or older and could walk household distances with or without an assistive device and without assistance from another person. Exclusion criteria included neuromuscular disorders, active cancer treatment, hospitalization for a life-threatening illness or major surgery in the last 6 months, severe pulmonary disease, chest pain with activity or a major cardiac event such as a heart attack or stroke in the past 6 months. Data was collected from the baseline testing session at the research and training laboratory of the Pittsburgh Pepper OAIC. Participation was offered to 216 older adults by telephone. Seventy-six older adults were not interested in participating and 20 were ineligible; the remaining 120 were enrolled. Seventy-eight of 120 participants had complete FOF, physical function and PA data and were included in this investigation (mean age = 77.6 (5.9)). Those excluded for missing data did not differ from those who were included in terms of age, gender, race or fall history (p<.05).

Measures

Demographic characteristics were collected by self-report and included age, gender and race. Fall history was defined as greater than one fall in the past year and was also determined by self-report. ⁹

Fear of Falling

We used both a global measure and an activity specific measure of self-reported fear status. The global measure of fear status, ‘..are you afraid of falling–yes or no’ was used to characterize the sample and provide information to enable readers to understand how the sample studied compared to the target population in prior reports. The SAFFE fear subscale was the measure of fear of falling chosen for the data analyses for two reasons: 1) the measure was designed to quantify fear related to physical activities in daily living, and 2) the scoring of the SAFFE fear subscale is a continuous variable, better suited for the relational analyses planned.

Self-reported fear status

Fear Status.⁹

A falls history questionnaire was used to determine current fear status from yes/no responses to the question, “Are you afraid of falling?”

Self-reported fear ratings

Survey of Activities and Fear of Falling in the Elderly (SAFFE.) – Fear Subscale.¹⁰

The SAFFE is an interview-administered instrument for measuring FOF related to basic and instrumental ADLs. Eleven activities of daily living are surveyed and three SAFFE subscales: fear, activity participation, and activity restriction determined from the responses. The SAFFE fear subscale score was determined based on the average fear ratings (0–3) across each of the 11 activities performed. For each of the 11 activities the format for the question is : “When you_________ [insert activity here, eg. go to the store], how worried are you that you might fall?” A score of “0” indicates no fear; “1” indicates “a little”; 2 indicates “somewhat” fearful; and 3 indicates “very” fearful while performing each respective activity. For each activity, the range of scores is 0–3. Internal consistency reliability and concurrent validity have been established. ¹⁰ SAFFE fear scores are significantly related to the Falls Efficacy Scale (FES) (−.76) and a one-item afraid of falling question (−.59).¹⁰

Physical Function

Late Life Function and Disability Instrument (LLFDI).¹¹

The self-reported LLFDI total functioning component was used to quantify physical functioning. The function component has 32 items in three areas: basic lower extremity function, advanced lower extremity function and upper extremity function. Each item is rated on a 0–5 scale. The raw scores were converted to scaled scores of overall function and were used to describe physical function in our sample. The LLFDI has a range of scores from 0–100, with higher scores indicating better function. Test-retest intraclass correlations have been reported as .91–.98.¹¹ LLFDI overall function scores are moderately associated with performance-based measures of physical function like the Short Physical Performance Battery (SPPB) (r=0.65, p<.001) and 400-meter walk gait speed (r=.69, p<.001).¹²

Physical Activity

Actigraph® Accelerometer^a

Participants’ usual daily physical activity was assessed using an Actigraph® accelerometer worn on the waist during waking hours for seven consecutive days. Accelerometry is a reliable method of assessing free-living physical activity¹³ and has been validated against direct observation,¹⁴ and energy expenditure.^{13, 15, 16} Written and verbal instructions were provided to the participant for the application and use of the Actigraph® accelerometer. Participants were instructed to remove the monitor only for sleeping, during imaging studies, and/or during swimming and bathing activities. Each participant noted the time the accelerometer was put on and removed daily in a diary. The mean daily counts of activity per minute worn were averaged over the seven day period.¹⁷

Exercise

SAFFE – Activity Participation Subscale – Walk for Exercise.¹⁰

The SAFFE Activity Participation subscale assess eleven different activities including getting out of bed, reaching for an object overhead, visiting family, and going to crowded places, in addition to walking for exercise. We used the walking for exercise item to represent exercise participation. Participants were asked “Do you currently take a walk for exercise?” Answers were recorded as yes or no.

Data analysis

Descriptive statistics (mean, SD and range) were calculated for age, demographic characteristics and the fear, physical function and activity measures. The relation of FOF and PA was calculated using Pearson correlations among FOF (SAFFE Fear), physical activity and physical function. In order to explore relationships between exercise, physical function and PA, we stratified the sample by subsets of older adults based on exercise and based on physical function groups. Analyses were conducted for the entire sample and for the sample stratified by walk for exercise status and stratified by higher (non-mobility limited, LLFDI ≥ 59.8) and lower functioning (below non-mobility limited, LLFDI < 59.8) groups. Linear regression was used to define the relation of FOF (SAFFE Fear) to physical activity, controlling for age, gender and physical function.

RESULTS

The mean age of the older adults studied was 77.6 years, and almost three-quarters of the subjects were female. About 40% of the subjects had fallen and a similar proportion reported fear of falling. The mean SAFFE fear scores among those who were afraid of falling using the yes/no question was .65; among those who were not afraid of falling the mean SAFFE fear score was .22 (f=24.5, p<.001). Physical function (LLFDI) was generally representative of healthy, independent adults (Table 1).

Table 1.

Subject characteristics (n=77)

Subject Characteristics	Total	Afraid of Falling	Not Afraid of Falling
Age (y)	77.4±5.8	77.7±5.7	77.0±6.1
Female	58 (75.3)	32 (55.2)	26 (44.8)
Male	19 (24.7)	4 (21.1)	15 (78.9)
Black	10 (12.8)	2 (22.3)	7 (77.8)
Education ≥13y	54 (70.1)	23 (42.6)	31 (57.4)
Fallen in the past year	31 (40.3)	15 (48.4)	16 (51.6)
FOF/fear status
SAFFE fear	0.39±0.42	0.64±0.50	0.22±0.25
Fear status—yes	31 (39.7)
Physical function
LLFDI function	60.3±9.7	55.4±8.7	63.5±9.1
PA
Accelerometer, cpm/d	148.5±77.9	143.0±61.9	152.1±87.0
Walk for exercise—yes	52 (67.5)	33 (63.5)	19 (36.5)

NOTE. Values are mean ± SD or n (%).

Abbreviation: cpm/d, counts per minute per day.

In the total sample, function, fear and physical activity were related, although the relation between fear and physical activity was weak (p=.02) (Table 2). When stratified by exercise status, fear was no longer related to physical activity (p=.17 and .31, respectively), but the relation between fear and physical function remained (p<.001) (Figure 1a and b). When stratified into higher and lower functioning groups, fear was not related to total daily physical activity (p=.07 and p=.70, respectively) (Table 2).

Table 2.

Relations between age, FOF, PA, and physical function, stratified by exercise and functional status

Variable	Age	PA	SAFFE Fear
Total (N=78)
Age
PA	−.39 (<.001)
SAFFE fear	.01 (.930)	−.26 (.020)
LLFDI total function	−.13 (.270)	.45 (<.001)	.61 (<.001)
Walk for exercise (n=52)*
Age
PA	−.45 (.001)
SAFFE fear	−.19 (.170)	−.19 (.170)
LLFDI total function	−.02 (.870)	.38 (.006)	−.56 (<.001)
Don’t walk for exercise (n=25)*
Age
PA	−.28 (.170)
SAFFE fear	.23 (.260)	−.21 (.310)
LLFDI total function	−.39 (.060)	.11 (.590)	−.72 (<.001)
LLFDI higher functioning (n=40)†			NA
Age
PA	−.29 (.070)
SAFFE fear	−.17 (.300)	−.29 (.070)
LLFDI lower functioning (n=38)‡			NA
Age
PA	−.48 (.002)
SAFFE fear	−.01 (.940)	−.06 (.700)

NOTE. Values are r (P).

Abbreviation: NA, not applicable.

^*SAFFE activity participation subscale, n=77.

^†Nonmobility limited, LLFDI ≥59.8.

^‡Below nonmobility limited, LLFDI <59.8.

Figure 1.

a. Relations of fear of falling, physical activity and function for total sample..

b. Relations of fear of falling, physical activity and function for groups defined by exercise status.

SAFFE: Survey of Activities and Fear of Falling in the Elderly; LLFDI: Late Life Function and Disability Instrument.

*p<.01

**Based on yes/no response to SAFFE activity scale question “Do you currently take a walk for exercise?”

Results of the linear regression show that physical function explained 19% of the additional variance in physical activity when controlling for age and gender (Table 3, Model 2, R² Change = .190). The addition of fear status to the model did not explain any more of the variance in physical activity (Table 3, Model 3, R² Change = .000).

Table 3.

Contribution of physical function and fear on PA

Variable	β	P	R2	R2 Change
Model 1			.158	NA
Age	−.374	.001
Sex	.150	.161
Model 2			.344	.190
Age	−.299	.002
Sex	.247	.012
Physical function	.450	<.001
Model 3			.335	.000
Age	−.300	.003
Sex	.247	.012
Physical function	.444	<.001
Fear	−.010	.931

Abbreviation: NA, not applicable.

DISCUSSION

Fear of falling and total amount of daily activity were weakly related in this sample, and the relation no longer existed when groups were stratified by either exercise status or functional status. While this was an unexpected finding, the implications may be important. Self-reported or performance-based measures of function describe a person’s capacity, while accelerometry used to record activity over a period of time measures what they actually do. Fear of falling has the potential to limit both capacity and daily activity accumulation. In our study, we expected that those who were fearful of falling would have both lower capacity for activity and lower total daily activity accumulated. However, we found that fear and physical activity were weakly related at best and unrelated when the sample was stratified by exercise or functional status. In our sample, some people accumulated physical activity despite the presence of fear. Clinically, this is important because participating in activity despite being fearful may reinforce the fear of falling and thus contribute to anxiety or an unwillingness to advance activity beyond the “must-do” activities of daily life. This form of fear associated with anxiety may be treatable using a variety of techniques that have been described in the literature.^{18, 19}

In the presence of moderate or better relations between fear of falling and function, the weak relation between fear of falling and physical activity might be explained several ways. Persons who are fearful of falling have been known to restrict the number of activities they perform, some to a severe extent such that their behavior resembles an anxiety disorder such as agorophobia.²⁰ However, the people who restrict physical activity because of fear of falling may still accumulate activity that is similar to their non-fearful counterparts because much of the activity performed by older adults is known to be low-intensity, ADL-type activity. ²¹ So while some people may be “paralyzed by fear” and not accumulate much activity at all, others perform certain ADLs and IADLs because the daily activities must be done: a person must walk to the bathroom lest they become incontinent; a person must grocery shop and prepare a meal because they have no one to assist them with this task, even if they are fearful. As an alternative explanation, some people may increase their activity if their fear of falling is associated with an anxiety characterized by pacing or repetition of activities. Based on a mean SAFFE score of .39, our sample reflects the performance of people with low levels of fear. Among people with more pronounced fear of falling, future studies may explore the relation between anxiety, function and physical activity.

Some people may continue to exercise despite a reported fear of falling, as evidenced by our results that 66% of those who were afraid of falling report that they “take a walk for exercise.” This is consistent with a study by Shimada, et al. who reported that 50% of older adults who were afraid of falling continued regular physical activity over a 2 year period despite reported fear.²² The older adults may believe exercise is beneficial for their health and thus they may exercise despite reported fear in order to reap the health benefits.²¹

Based on our results, fear of falling was not related to total daily activity when the subjects were stratified by exercise status. We expected that those who did not walk for exercise would have greater levels of fear, but in our sample this did not prove to be true. Regardless of the levels of FOF, some people still walk for exercise. Brach, et al,⁶ conducted a study to determine whether older adults who exercise had higher levels of physical function than those whose intensity of daily activity was categorized as either “lifestyle inactive” or “lifestyle active” based on self-reported energy expenditure. In that study, both exercise and lifestyle activity were shown to be beneficial in terms of function, with the exercise group demonstrating better function than the “lifestyle active” group. This indicates that the intensity of physical activity, whether during exercise or during functional tasks, is an important factor in functional performance. We did not determine the intensity of the daily activity performed in this study. While our results show that FOF is not related to exercise status, FOF may in fact limit the intensity of exercise performed, thereby reducing the functional benefits derived from these activities. Future studies should examine the intensity of physical activity among those who are fearful of falling compared to those who are not.

People who are fearful of falling may demonstrate or report the capacity for adequate or good physical function, but may not accumulate physical activity that is equal to their capacity. For those people it may be clinically important to use daily functional tasks as interventions, instead of traditional exercise or walking programs, to increase activity levels to match or exceed their functional capacity in order to maintain or improve health.

Study Limitations

The SAFFE fear scores reported (.39) appear to be low compared to other studies of community-dwelling older adults.^{10, 23} Howland and Lachman have previously described fear of falling levels as reported on the SAFFE of less than .40 to be consistent with “not afraid of falling”.^{10, 23} Although the level of fear in this sample is low, it is still related to function, and still may influence daily activity. Despite the low mean SAFFE fear score, 40% of the participants answered “yes” to the general question “Are you afraid of falling?”

Activity-specific measures used to assess fear related to a specific task, like the SAFFE, may not assess fear in all circumstances that can be fear-producing. The functional status of this sample represents healthy, independent older adults who are not mobility-limited (Mean LLFDI score = 60.12).¹² As such, they may be fearful of some higher-level activities that are not represented by the SAFFE. For example, the SAFFE does not address fear related to walking up or down stairs, a common functional task that may evoke fear in older persons. In these cases, older persons may be fearful while still accumulating daily activity. It is also possible that people who are fearful may report fear indirectly, as a limitation in function. The LLFDI function subscale requires respondents to report how much difficulty they have performing a specific task. The task may be difficult due to fear, but a person may continue to do the task because it is a necessary daily activity. In that case, a fearful person and a non-fearful person would accumulate the same amount of activity.

Lastly, participants may have increased their daily activity as a response to wearing the accelerometer. The accelerometer has no external display; therefore, the participants are not able to view any representation of their activity level. Despite the lack of visual feedback, participants may have increased their activity level because they knew activity was being recorded.

CONCLUSION

Fear of falling was related to physical function and only weakly related to daily physical activity in this group of community-dwelling older adults. When stratified by exercise and functional level, the relation of FOF to physical activity no longer existed. Self-reported physical function may describe a person’s capacity to perform an activity, while recorded PA reflects what they actually do. Daily activity may include things that a person “must do” despite reported fear. Clinicians who work with community-dwelling older adults should recognize that older adults may indeed be fearful even if their amount of daily activity resembles that of their non-fearful counterparts. A thorough assessment of FOF status is needed, including the possibility that fear may be reported as a limitation in function. Strategies to reduce fear and increase confidence may be beneficial in order to improve function.

ABBREVIATIONS

FOF

Fear of Falling

PA

Physical Activity

SAFFE

Survey of Activities and Fear of Falling in the Elderly

LLFDI

Late Life Function and Disability Instrument

SPPB

Short Physical Performance Battery