Falls are a dangerous and deadly epidemic among seniors. They are the leading cause of accidental death in older adults and result in heavy health care services use and premature nursing home admissions (Centers for Disease Control and Prevention [CDC], 2010). For the purposes of this article, a fall is “an event which results in a person coming to rest inadvertently on the ground or floor or other lower level” (World Health Organization, 2010, p. 1). Nationally, one-third of adults >65 yr old and nearly half of those >80 yr old fall every year (CDC, 2009). Of those who fall, 20%–30% incur moderate to severe injuries that may interfere with mobility and independence and that increase their chances of early death (CDC, 2008).

Fear of falling, a psychological symptom, is prevalent in 35%–55% of community-dwelling older adults, regardless of whether they have experienced a fall (Boyd & Stevens, 2009; Delbaere, Crombez, Van Den Noortgate, Willems, & Cambier, 2006; Zijlstra et al., 2007). Fear of falling is defined as a “lasting concern about falling that can lead an individual to avoid activities that he/she remains capable of performing” (Tinetti, Richman, & Powell, 1990, p. 239). Multifaceted factors contributing to fear of falling have been independently associated with women, multiple medications (≥4), fall history, fall-related injuries, age, decreased activity of daily living (ADL) performance, activity restriction, physical deconditioning (e.g., lowered function, strength, balance, mobility), social isolation, decreased quality of life, and sedentary lifestyle (CDC, 2009).

Past research has demonstrated that 20%–60% of older adults limit their activities because of fear of falling (Deshpande et al., 2008; Hadjistavropoulos et al., 2007; Jang, Cho, Oh, Lee, & Baik, 2007). Activity restriction, in itself, has been found to be a fall risk factor because it can lead to muscle weakness, poor balance, deconditioning, and social isolation (Gagnon & Flint, 2003). Jang and associates (2007)  determined that older adults who experienced a fall within the past 6 mo were most likely to develop fear of falling and had a fivefold likelihood of restricting their activities compared with older adults who had not fallen. The degree of fear of falling has been associated with the intensity of depression and anxiety and with self-efficacy, which directly influences activity engagement (Delbaere, Close, Brodaty, Sachdev, & Lord, 2010; Gagnon, Flint, Naglie, & Devins, 2005; Jang et al., 2007; van Haastregt, Zijlstra, van Rossum, van Eijk, & Kempen, 2008). Few studies, however, have examined the clinical consequences of fear-related activity restriction and whether and how such activity restriction is associated with depression and anxiety (Arfken, Lach, Birge, & Miller, 1994; Delbaere et al., 2006; Fletcher, Guthrie, Berg, & Hirdes, 2010; Gagnon et al., 2005; Hadjistavropoulos et al., 2007).

Depression is the most common mental health disorder among older adults (12%–20%; Porzych, Kadziora-Kornatowska, Porzych, Polak, & Motyl, 2005), and older adults who acknowledge fear of falling have the highest level of depression (37.5%; Drozdick & Edelstein, 2001; Gagnon et al., 2005). Although fear of falling was not a focus of a study by Anaby, Miller, Eng, Jarus, and Noreau (2009), their findings suggest that the degree of depression accompanied by level of physical disability determined older adults’ activity level or restriction.

Anxiety is frequently found among both depressed and nondepressed community-dwelling older adults (Mehta et al., 2003). Past research has found that 8%–18% of community-dwelling older adults experience anxiety, that anxiety occurs in about 50% of those with depression, and that anxiety affects physical and psychosocial functions (Drozdick & Edelstein, 2001; Gagnon et al., 2005; Mehta et al., 2010; Porzych et al., 2005). Gagnon and associates (2005)  suggested that anxiety disorders have a significant independent association with fear of falling.

Fear of falling may create a vicious spiral of downward physical and occupational performance and activity restriction. Because activity engagement and occupational performance are central tenets of occupational therapy, it is imperative that occupational therapy practitioners screen their older adult clients for fall risk, fear of falling, activity restriction, depression, and anxiety (AOTA, 2008). No consensus among health care providers exists, however, as to which screening tools are most appropriate in the community setting to ascertain whether older adults are at risk for falling, fear of falling, depression, anxiety, and accompanied activity level.

With the expected increase in the population of adults >65 yr old, health care systems and community, local, state, and federal agencies are anticipating an increase in the need for services for older adults who have fallen and display self-imposed activity restriction, fear of falling, anxiety, or depression (Administration on Aging, 2007). Thus, it is imperative that health care providers prepare for this escalating problem of falls, fear of falling, and associated consequences among older adults.

The correlational and nonexperimental study described in this article extends the work of Lachman and associates (1998)  investigating activity levels associated with fear of falling when using the Survey of Activities and Fear of Falling in the Elderly (SAFE) assessment. We used the SAFE, the Geriatric Depression Scale–30 (GDS; Kieffer & Reese, 2002), and the Hamilton Anxiety Scale, IVR Version (HAMA; Hamilton, 1969), for this study. Our first objective was to relate each participant’s fear of falling score from the SAFE to depression and anxiety scores, SAFE activity level score, and SAFE activity restriction score. Our second objective was to determine whether depression and anxiety were predictive of each SAFE item score.

We collected data from participants in seven community centers and one apartment complex in a rural eastern North Carolina county. A convenience sample using purposive and snowball sampling was used. Community-dwelling adults ≥55 yr old who had or had not experienced falls were invited to participate in the study. Adults were excluded from the study if they did not speak English or had a cognitive or developmental disability. We initially notified participants about the study through fliers posted at the community centers and apartment complex community bulletin board. The flier identified the name of the study and when and where the researchers would meet interested older adults who wished to learn more about the study.

The researcher (Jane A. Painter) and three graduate occupational therapy students (Puneet Dhingra, Justin Daughtery, and Kira Cogdill) met with the older adults in their own environment (i.e., community center or apartment complex) and described the purpose of the study and their role in the study. The researcher and students explained the details of scheduling the one-time, 2-h appointment; invited potential participants to a 90-min fall prevention presentation; and described the sensor night-light gift that study participants would receive on completing their part. The 90-min fall prevention presentation reviewed fall risk factors, ramifications of falls and fear of falling, home safety strategies, and community resources. A grant-funded community organization provided the sensor night lights to all participants who completed the study and attended the presentation. To prevent exclusion from the study because of illiteracy or visual, hearing, or other deficits, each participant was given the opportunity to independently complete the SAFE, GDS, and HAMA or to have the assessments read to him or her. All assessments were administered in the same sequence. Because of time constraints, we did not debrief the participants. We obtained informed consent from each participant before data collection. The institutional review board of East Carolina University approved the study.

We used an adapted, semistructured fall questionnaire (Berkman & Miller, 1991) to collect specific information regarding participants’ fall history and consequences of identified falls within the past 6 mo. Although the developers of this questionnaire did not determine validity or reliability, they piloted the questionnaire several times and made modifications, and findings suggest that the survey provides practitioners with appropriate fall history information (Berkman & Miller, 1991). All participants completed Part 1 of the questionnaire, which comprises five questions requesting demographic information on gender, age, living status, and working status (including volunteer and paid work) and asking whether the respondent had experienced a fall or near fall (Berkman & Miller, 1991). Part 2 of the questionnaire, comprising 16 questions, was designed to be administered only to those who had fallen and focuses on the number of falls experienced, specific information regarding the last fall, and use of ambulatory devices and safety equipment (Berkman & Miller, 1991). Questions include a combination of yes-or-no and multiple-choice responses.

The GDS long form is a 30-item valid (α = .94) and reliable (r = .85) yes-or-no screening tool that has been used across cultures with adults >65 yr old (Kieffer & Reese, 2002; Yesavage et al., 1982). GDS scores of 0–9 are considered normal, 10–19 suggest mild depression, and ≥20 indicate severe depression (Yesavage et al., 1982).

The HAMA, a valid (α = .92) and reliable (r = .93) screening tool, is a widely used and accepted outcome measure for the evaluation of anxiety in various settings. It measures physical, somatic, and behavioral symptoms through 14 items with responses on a 5-point scale. Scores indicate no (0), mild (1–18), moderate (19–25), and severe (26–30) levels of anxiety (Beck, Stanley, & Zebb, 1999; Hamilton, 1969; Maier, Buller, Philipp, & Heuser, 1988).

The SAFE, a reliable (α = .95) and valid (r = .91) instrument, assesses fear of falling, activity restriction, and activity level (Lachman et al., 1998). The SAFE assesses 11 activities. Community-based activities consist of shopping, walking outside home, walking on a slippery surface, visiting relatives or friends, going to a crowded place, and going for long walks. Home-based activities include cooking, taking a tub bath, getting out of bed, and reaching up or bending down to pick up objects. For each of the 11 activities, participants respond to six items, as follows:

• SAFE A: Do you currently do the activity? (yes or no)

• SAFE B: If you do the activity, to what degree are you worried you might fall? (0 = not at all worried, 1 = a little worried, 2 = somewhat worried, and 3 = very worried)

• SAFE C: If you don’t do the activity, do you not do it because you are worried you might fall? (0 = not at all worried, 1 = a little worried, 2 = somewhat worried, and 3 = very worried)

• SAFE D: If you do not do the activity because of worry, are there also other reasons you do not do it? (if yes, specify reason)

• SAFE E: If you are not worried, what are the reasons you do not do the activity? (specify)

• SAFE F: Compared with 5 yr ago, would you say you do the activity 1 = more than you used to, 2 = about the same, or 3 = less than you used to?

A higher SAFE B score indicates higher fear of falling, whereas a higher SAFE A score indicates higher activity level. Note that no score is calculated for SAFE E; rather, it gives the person an opportunity to describe other reasons an activity is not completed after answering the SAFE C question.

Like other, better known fear-of-falling scales, the SAFE inquires about fear experienced during performance of several functional items. Unique to the SAFE is measurement of the extent to which the respondent actually performs or restricts the activity (i.e., activity level and activity restriction measures). Moreover, when activity restriction exists, the SAFE delineates the degree to which fear of falling is the causative factor leading to self-restriction.

To ensure confidentiality, participants did not include any identifiable personal information, and all instruments were assigned a numeric code. Data from the fall demographic questionnaire, SAFE, GDS, and HAMA were collected by the researcher and three graduate occupational therapy research students at the participants’ community center or apartment. During data collection, participants were seen individually in a private room to ensure confidentiality. The occupational therapy graduate students were trained in the procedures for each assessment and protocol for data collection to ensure interrater reliability.

Descriptive statistics were used to profile the participants’ characteristics. As an initial step, the relationships among fear of falling, depression, anxiety, activity level, and activity restriction were established using Spearman’s ρ (nonparametric data). Subsequently, five separate regression analyses were used to determine whether depression and anxiety predicted SAFE item variables. Independent variables in all analyses were depression and anxiety. Dependent SAFE variables included (1) activity level, (2) fear of falling, (3) activities not done for reasons other than fear of falling, (4) activities not done for reasons in addition to fear of falling, and (5) activity restriction (total). SPSS Version 15 (SPSS, Inc., Chicago) and a significance level of p ≤ .05 were used for all analyses.

A total of 99 people—84 from senior centers and 15 from a senior apartment complex—matched the eligibility criteria and were included in this study. Table 1 displays the characteristics of the research sample. Collectively, 38.4% reported a fear of falling, as measured by SAFE B.

Fear of falling, as measured by SAFE B, was significantly correlated with (1) depression, as measured by the GDS; (2) anxiety, as measured by the HAMA; and (3) activity level, as measured by SAFE A (r = −0.210, p < .037; see Table 2). Depression, as measured by the GDS, was significantly correlated with anxiety, as measured by the HAMA. Activity restriction, as measured by SAFE F, was significantly correlated with depression, as measured by the GDS. Activity level, as measured by SAFE A, was significantly negatively correlated with activity restriction, as measured by SAFE F; fear of falling, as measured by SAFE B; depression, as measured by the GDS; and anxiety, as measured by the HAMA see Table 2).

As shown in Table 3, activity restriction because of fear of falling, as measured by SAFE B, as well as for other reasons, as measured by SAFE D, was predicted by both depression, as measured by the GDS, and anxiety, as measured by the HAMA. Activity level, as measured by SAFE A, was predicted by anxiety, as measured by the HAMA, but not by depression, as measured by the GDS. Fear of falling, as measured by SAFE B, was predicted by anxiety, as measured by the HAMA, but not by depression, as measured by the GDS. Activities not performed for reasons other than fear of falling, as measured by SAFE C, were not predicted either by depression, as measured by the GDS, or by anxiety, as measured by the HAMA. Activity restriction, as measured by SAFE F, was not predicted either by depression, as measured by the GDS, or anxiety, as measured by the HAMA.

Clinical implications drawn from previous studies (Deshpande et al., 2008; Fletcher et al., 2010; Jang et al., 2007) indicate the need for occupational therapy practitioners to screen older adults for fall risk, fear of falling, and their associated ramifications, including activity restriction, anxiety, and depression. Findings from the current study support this view. Our sample, though not randomly selected, represents the national average of one-third of older adults who have sustained a fall. Thirty-eight percent of our convenience sample reported experiencing fear of falling, within the range reported in past research (Boyd & Stevens, 2009; Delbaere et al., 2006; Zijlstra et al., 2007).

A strength of the current research relates to using the SAFE assessment, which does not require participants to estimate performance of activities they do not do (Lachman et al., 1998). Past research has suggested that using an actual activity restriction measurement rather than asking participants to estimate performance is far better in determining the effect fear of falling has on activity engagement (Gagnon & Flint, 2003; Lachman et al., 1998; Li, Fisher, Harmer, McAuley, & Wilson 2003). Of the 11 activities identified in the SAFE instrument, 88% of the participants restricted their activity level by at least one activity, and 49% indicated, compared with 5 yr previously, that they were now engaging in fewer activities. The proportion of participants restricting their activity by at least one activity is higher than past findings, but the proportion of participants engaging in fewer activities than 5 yr ago is comparable to past research (Deshpande et al., 2008; Fletcher et al., 2010; Jang et al., 2007). The significant relationship between anxiety and depression was comparable to past studies (Delbaere et al., 2010; Gagnon et al., 2005; Porzych et al., 2005).

Findings from the current study offer health care providers new insights regarding the interrelationship between fear of falling and activity level, fear of falling and depression, and fear of falling and anxiety. Current results suggest the importance of addressing both anxiety and depression in relation to fear of falling and activity level. Unlike past research (Gagnon et al., 2005; van Haastregt et al., 2008), the current study found that anxiety had a stronger relationship to fear of falling than did depression. Interestingly, past research (Anaby et al., 2009; Arfken et al., 1994) determined a strong association between depression and fear of falling but neglected to simultaneously examine the relationship between anxiety and fear of falling, even though depression and anxiety frequently occur simultaneously among older adults (Mehta et al., 2003, 2010; Porzych et al., 2005). Moreover, many studies (Delbaere, Crombez, Vanderstraeten, Willems, & Cambier, 2004; Deshpande et al., 2008; Hadjistavropoulos et al., 2007; Yardley & Smith, 2002; Zijlstra et al., 2007) failed to address how anxiety and depression affect fear of falling and activity restriction.

Our findings, coupled with those of past research (Delbaere et al., 2010; Drozdick & Edelstein, 2001; van Haastregt et al., 2008), suggest that older adults who were more actively engaged in ADLs and instrumental activities of daily living had lower levels of fear of falling, depression, and anxiety. Conversely, participants who were more fearful, depressed, and anxious had restricted activity levels. Our study found that anxiety was the only independent variable that predicted fear of falling, activity level, and number of activities the person currently performed compared with 5 yr previously, suggesting, like Mehta and colleagues’ (2010) findings, that self-efficacy influences a person’s perceptions of capabilities and activity level to a high degree.

Unlike past studies (Drozdick & Edelstein, 2001; Gagnon et al., 2005; van Haastregt et al., 2008), our study found that depression did not predict fear of falling, activity level, or activity restriction. How might this discrepancy be explained? Compared with the depression scores of our participants, anxiety scores were well distributed across the mild to severe levels of anxiety. Conversely, only 1 participant had severe depression, and 10 had mild depression. A lack of depressive symptoms in the current sample may explain why depression did not predict fear of falling, anxiety, or activity level. The relatively low levels of depression found in this convenience sample may indicate that our participants enjoyed a higher level of social and activity engagement than might have been found in a random sample. We recruited most participants from community centers where, presumably, they had self-selected to participate in social contact and activity programs.

In summary, occupational therapy practice implications include the following considerations:

• A person can develop a fear of falling whether or not a fall has occurred.

• Anxiety has been found to cause a fear of falling.

• Fear of falling can lead to increased fall risk, decreased motivation, and decreased perceptions of capabilities, all of which can then lead to self-imposed activity restriction.

• Occupational therapy practitioners should assess for fear of falling and anxiety for older adults who display anxiety and decreased motivation to perform functional activities.

Our study has several limitations. First, because we used a small convenience sample, results may not be generalized to larger populations of community-dwelling older adults. Second, our sample could be considered heterogeneous in that participants were all community-dwelling older adults. Future studies could examine a more diverse sample of older adults of varying health status and living environments. Third, the data we collected were self-reported and dependent on participants’ memories of past events, with the attendant risk of underestimation of falls, fear of falling, anxiety, depression, activity level, and activity restriction. Future studies should include objective measures to better understand the multifaceted construct of fear of falling.

Occupational therapy practitioners may better tailor their interventions to meet clients’ individualized needs if they have an improved understanding of the relationship between activity restriction related to fear of falling and the consequences of anxiety and depression. This study found a relationship between (1) fear of falling and anxiety, depression, and activity level; (2) depression and anxiety; (3) activity restriction and depression; and (4) activity level and activity restriction. Activity level was negatively correlated with activity restriction, fear of falling, depression, and anxiety. Depression and anxiety predicted activity restriction because of fear of falling and for other reasons. In addition, anxiety predicted both activity level and fear of falling. Because activity engagement and occupational performance are central tenets of occupational therapy (AOTA, 2008), it is of paramount importance that occupational therapy practitioners screen their older adult clients for fear of falling, anxiety, and depression because these factors may lead to fall risk and activity restriction. Moreover, fear of falling may occur regardless of whether a person has actually fallen, because it results from the person’s self-perceived notion of occupational function. Thus, a self-fulfilling prophecy may occur in which fear of falling places the person at a higher risk of falling and lowers his or her degree of motivation to engage in activities.

Comparing our results to past research is a daunting task because of the great variation in measurement tools used to investigate fear of falling, depression, anxiety, activity level, and restriction. In addition, because fear of falling is so complex, past studies have investigated different variables, in different combinations and environments, to determine its causes and ramifications. Several possible areas for future research are suggested by findings from the current study. First, development of a single reliable and valid instrument that would measure all three variables—fear of falling, depression, and anxiety—in community-dwelling older adults would decrease confusion in comparing future research investigating the reciprocal influence of fear of falling on depression and anxiety. Second, determining which existing fear of falling instruments are the most appropriate to use in varying living and health care environments would also assist in comparative studies. For example, should the same instrument be used for a healthy community-dwelling older adult and a frail person who lives in an assisted-living or skilled-nursing facility? Occupational therapy practitioners could examine how the degree of fear of falling, anxiety, and depression affects activity engagement in older adults in varied living and health care environments and of varied health status levels.

In conclusion, findings from our study suggest that when developing appropriate individualized interventions, occupational therapy practitioners should assess how fear of falling (regardless of whether a fall has occurred), anxiety, and depression affect their older adult clients’ occupational performance and activity engagement.

The study was supported in part by Pitt Partners for Health, Older Adult Health Subcommittee, funded through Pitt County Memorial Hospital Foundation.