Falls Risks and Prevention Behaviors Among Community-Dwelling Homebound and Non-Homebound Older Adults

Abstract

The objectives were to examine falls risk factors to determine how the magnitude of risk may differ between homebound and non-homebound older adults, and to describe falls prevention behaviors and participation in falls prevention education. A cross-sectional survey was conducted with convenience samples of community-dwelling older adults recruited through Meals on Wheels programs (homebound, n = 80) and senior centers (non-homebound, n = 84) in North Carolina. Data were collected during home visits and included an interview and medication inventory. Multivariate negative binomial regression with robust variance estimation modeled risk factors for falls. Risk factors for falls observed in both the homebound and non-homebound populations are consistent with what is known in the literature. However, the magnitude of the risk was higher in the homebound than in the non-homebound population with respect to vision impairments, number of high-risk and over-the-counter medications, and use of walking aids .Few participants reported participating in a falls prevention program.

Keywords

older adults homebound falls prevention non-homebound cross-sectional survey

Introduction

Falls are the leading causes of unintentional fatal and nonfatal injury among older adults in the United States (National Center for Injury Prevention and Control [NCIPC], 2014). More than one third of older adults fall each year, and many of these falls result in decreased or permanent loss of functioning, loss of independence, and development of a fear of falling (Stevens, Mack, Paulozzi, & Ballesteros, 2008). Risk factors for falls among community-dwelling older adults are well established (Deandrea et al., 2010; Nevitt, Cummings, Kidd, & Black, 1989; Tinetti, Speechly, & Ginter, 1988). The strongest predictors of a fall include lower extremity weakness, history of falling, gait and balance deficits, and use of psychotropic medications (Rubenstein & Josephson, 2006). However, the effect of these factors on falls risk has not been widely examined for homebound subgroups of community-dwelling older adults.

Homebound older adults tend to be older and have more physical and mental impairments than their non-homebound counterparts (Herr, Latouche, & Ankri, 2013; Qui et al., 2010). Increasing age, impaired activities of daily living, cognitive impairments, and poor balance and gait are all associated with falls (Ganz, Bao, Shekelle, & Rubenstein, 2007; Nevitt et al., 1989; Stenhagen, Ekström, Nordell, & Elmståhl, 2013). It is therefore expected that homebound older adults will be at higher risk for falling in part because they are frailer and may not be receiving sufficient caregiving or home care assistance (Danilovich, Corcos, Marquez, Eisenstein, & Hughes, 2015; Hickam et al., 2013; Levine, Boal, & Boling, 2003).

Evidence drawn from systematic reviews, such as from Gillespie et al. (2012), indicate how falls among community-dwelling older adults can be prevented (Gillespie et al., 2012). Effective falls prevention programs include home- and group-based exercise that address balance and strength, Tai Chi, and multifactorial programs that include exercise as well as recommendations for medication reviews, footwear, vision checks, and home safety (American Geriatrics Society/British Geriatrics Society [AGS/BGS], 2011; Gillespie et al., 2012). However, the efficacy of most of these programs has been evaluated among ambulatory populations. There is some evidence that exercise programs composed of strength, endurance, and balance training can reduce falls among frail, community-dwelling older adults (Cadore, Rodriguez-Manas, Sinclair, & Izquierdo, 2013). The effectiveness of multifactorial programs in reducing falls among frail, community-dwelling older adults is mixed (de Vries et al., 2010; Faes et al., 2011; Gillespie et al., 2012; Hendriks et al., 2008).

The prevalence of older adult engagement and adherence to falls prevention recommendations vary by the type of program and recommendation. Adherence rates range from 64.2% for exercise programs to 68.4% for multifactorial falls prevention programs, but from 28% to 95% for the individual components of multifactorial programs (Nyman & Victor, 2012). In studies examining older adults’ perspectives of fall risks, many feel that falls information is not relevant to them, whereas others are afraid of falling or fearful of losing their independence if they did fall (McMahon, Talley, & Wyman, 2011; Stevens, Noonan, & Rubenstein, 2010). These fears, however, do not necessarily translate into older adults’ participation in falls prevention programs. Barriers to participation have included perceptions that falls are not preventable, being unaware of the risk for falling, feeling limited by health and physical function, not feeling supported by health care providers and family, and not having access to programming (Bunn, Dickenson, Barnett-Page, McInnes, & Horton, 2008; McMahon, Talley, & Wyman, 2011). Although a considerable amount is understood about engagement in falls prevention behaviors and barriers to program participation, little is known about how this may differ between community-dwelling homebound and non-homebound older adults.

The objectives of this article were to (a) examine risk factors for falls among homebound and non-homebound older adults to determine how the magnitude of risk may differ between the two populations, (b) describe falls prevention behaviors among homebound and non-homebound older adults, and (c) describe participation in falls prevention education among homebound and non-homebound older adults.

Method

Study Design and Population

This study was a cross-sectional survey of community-dwelling older adults living in rural and urban counties in central North Carolina. To be eligible for the study, older adults had to be at least 65 years of age, non-wheelchair bound, and able to read and write in English. Older adults were recruited through four Meals on Wheels (MOW) programs (homebound population) and five senior centers (non-homebound population). Homebound status was defined as receiving MOW services at the time of study recruitment (Houston et al., 2015). The study protocol was approved by the University of North Carolina Institutional Review Board, Number 11-1237.

Homebound older adults were recruited by research staff who rode with MOW volunteers on their food routes. The research staff member gave each older adult food recipient a study brochure and provided an overview of the study. The age of the MOW recipient was determined using a simple screener at the time of food delivery. If age eligible and interested in participating in the study, the research staff member recorded their contact information and called the older adult at a later time to schedule a date to complete consent procedures and data collection. As the prevalence of cognitive impairments among homebound older adults can be high (Herr et al., 2013), each older adult was required to pass the six-item Mini-Mental State Exam (MMSE; Callahan, Unverzagt, Hui, Perkins, & Hendrie, 2002) prior to participating in the study. The MMSE was administered at the time of food delivery (if time permitted) or over the phone when the research staff member contacted the older adult.

A total of 367 homebound older adults were approached to participate in the study (Figure 1). Of those approached, 20 were excluded due to unsafe home environments for research staff to enter the homes, and 137 were not eligible to participate. The main reasons for non-eligibility were wheelchair bound (n = 47), below the age of 65 years (n = 40), and impaired cognition based on the MMSE (n = 35). Another 98 older adults approached were eligible to participate, of whom 80 (82%) enrolled. If an older adult was approached about the study but did not show interest in participating, eligibility was not assessed (n = 112).

Figure 1.

Recruitment and enrollment chart, homebound population.

Non-homebound older adults were recruited by research staff who set up booths in the five senior center lobbies. Recruitment at the senior centers occurred at the centers’ busiest times (normally during morning hours) to maximize the recruitment potential. Each recruitment event lasted approximately 3 hours. Senior centers were visited an average of 5 times depending on the size of the center. Older adults entering the centers were approached with a study brochure and an opportunity to learn more about the study. If older adults expressed interest in participating, they were screened for eligibility. If eligible, research staff scheduled a date and time for consent and data collection.

A total of 229 non-homebound older adults were approached to participate in the study (Figure 2). Of those approached, 26 (11%) were not eligible to participate, 97 (43%) were eligible, and eligibility was not assessed in the remaining 106 (46%). Among the 97 older adults who were eligible to participate, 84 (87%) enrolled.

Figure 2.

Recruitment and enrollment chart, non-homebound population.

Data Collection

Study data were collected during home visits by two trained research staff between March 2011 and September 2013. At the participants’ homes, research staff conducted an in-depth interview, a home safety assessment, and a medication inventory. The entire visit lasted an average of 1.5 hours, and participants were compensated US$50 for their time. The interview covered questions about falls risks, falls prevention behaviors and education, and demographics.

Falls risk

Data collected on falls risks included history of falls, fear of falling, medical conditions, confidence to maintain balance, home hazards, and medication. Falls history was collected as the total number of falls in the past year (AGS/BGS, 2011) and was defined as unintentionally coming to rest on the ground, floor, or other lower level. Participants were also asked to report their perceived fear of falling (Scheffer, Schuurmans, van Dijk, van der Hooft, & deRooij, 2008). Responses were coded on a 4-point Likert-type scale that ranged from very afraid to not at all afraid. In addition, participants were asked whether or not they had limited any activities due to a fear of falling, and if so, what specific activities were limited.

Medical conditions related to falls were assessed by asking participants to report problems with their vision, whether they required the assistance of a walking aid such as a cane or walker, and their overall physical and mental well-being. Participants were coded as having a vision impairment if they reported trouble with their vision when adjusting to changes in light, judging the steepness of stairs, or avoiding obstacles in their path. The Veterans RAND–12 (VR-12) validated self-assessment tool was used to collect overall physical and mental health. This tool is based on the Veterans RAND 36-Item Health Survey (VR-36), which was developed from the MOS RAND SF-36 Version 1.0 (Kazis et al., 2004). The survey tool is equally divided into eight physical and mental health domains. The questions are summarized into two overall measures of physical and mental health and weighted to a population average of 50. Scores below 50 represent physical or mental health scores below the population average.

Confidence to maintain balance was assessed using the 16-item validated Activities-specific Balance Confidence (ABC) scale developed by Powell and Myers (1995), which has been found to be predictive of falls among older adults (Lajoie & Gallagher, 2004). Participants were asked to rate their confidence in performing a number of activities related to daily living on a scale between 0% and 100%. Several participants were unable to provide an answer for some activities because they were not currently performing the activity. To account for these missing data, we calculated an overall ABC score by summing individual scores and dividing by the number of questions an individual answered. If a participant was missing more than half of the questions (7.3%), an overall score was not calculated. Higher ABC scores denote better confidence to maintain balance in performing functions related to daily living.

While one research staff member interviewed the participant, the second staff member conducted a home safety assessment and recorded participant medications. Development of the home safety assessment tool was guided by the Centers for Disease Control and Prevention document, “Check for Safety: A Home Fall Prevention Checklist for Older Adults” (NCIPC, 2005). The checklist covered the presence of handrails on both sides of stairways; objects blocking walking paths; damaged stairs, ramps, and carpet; lighting over stairways; anti-slip mats in the shower and bath; grab bars for the toilet and bath; light switches within reach of the bed; night lights between the bed and the bathroom; throw rugs without non-slip backing; and loose wires or cords on the floor. Following the walk-through, the research staff member asked the participant whether they use a step stool and whether there were any pets in the house. A home hazard score was calculated by summing the number of fall risk factors present in the house.

Participants were asked to provide the research staff with all prescription and over-the-counter medications they were currently taking. Medication name and whether the medication was prescribed or over-the-counter were recorded. High-risk prescription medications were defined as Benzodiazepines, Sedatives or Hypnotics, Antidepressants, Antipsychotics, Antihistamines, Opioids, and non-steroidal anti-inflammatory drugs (NSAIDs) because of their documented risk for falls among older adults (Hartikainen, Lönnroos, & Louhivuori, 2007; Woolcott et al., 2009).

Falls prevention behaviors and education

Participants were asked whether they were implementing the following falls prevention behaviors recommended by the AGS/BGS (2011): exercise on a regular basis, and if so, the type and frequency; annual vision checks; medication reviews by a physician or pharmacist to assess falls risk, and if so, the frequency; use of night lights in the bedroom and bathroom; and type of footwear typically worn.

Falls prevention education was assessed by asking participants whether they had ever taken part in a falls prevention program or whether they knew of a falls prevention program offered near their home. Participants were also asked whether they had ever been counseled about falls prevention by a health care provider, and if so, the type of recommendations that were given. Recommendations were reviewed by two research staff independently and grouped into themes; the overlap in these themes was compared for reliability.

Demographics

Age, gender, race, education, and living status were collected. Age was analyzed as a continuous variable, scaled 5 years. Living status was categorized as living alone or not living alone based on how many people participants reported living in their households.

Statistical Analysis

Chi-square or Wilcoxon’s tests and ANOVA models were used to examine associations of demographic and falls risk factors between the homebound and non-homebound populations. Fisher’s exact tests were used when variables had fewer than five participants in at least one category. Multivariate negative binomial regression models with robust variance estimation were used to identify risk factors associated with the number of falls in the previous year. Negative binomial models were used in place of Poisson due to violation of the assumption of equal variance. One set of models were adjusted for overall physical health using the results from the VR-12 physical health tool, and a second set were adjusted for age, race, and education to understand the independence of these risk factors with falls. Prevalence ratios (PRs) with 95% confidence intervals (CIs) were estimated for each risk factor. Statistical significance level was defined at α < .05. All analyses were carried out using SAS Version 9.3 (Cary, NC, USA) and were conducted starting in August 2014 after all data were collected.

Results

Study Population

The mean age of the participants was 77 years (SD = 8.0), and most were female (76%), White (56%), and had more than a high school education (56%) (Table 1). Homebound participants were significantly older, more likely to be Black, and have less than a high school education, compared with non-homebound participants. More than half of the participants reported living alone (59%), with a slightly larger percentage of homebound participants living alone compared with non-homebound participants.

Table 1.

Characteristics of Homebound and Non-Homebound Older Adult Populations.

Risk factor	Totaln = 164	Homeboundn = 80	Non-homeboundn = 84	p value
Demographics
Age (M ± SD)	77.4 ± 8.0	79.6 ± 8.1	75.3 ± 7.3	<.001
Gender, n (%)
Male	40 (24.4)	21 (26.3)	19 (22.6)	.590
Female	124 (75.6)	59 (73.8)	65 (77.4)	.590
Race, n (%)
White	91 (55.5)	31 (38.8)	60 (71.4)	<.001
Black	71 (43.3)	49 (61.3)	22 (26.2)
Other	2 (1.2)	0	2 (2.4)
Education, n (%)
<High school	13 (8.0)	11 (13.9)	2 (2.4)	<.001
High school	59 (36.2)	37 (46.8)	22 (26.2)
>High school	91 (55.8)	31 (39.2)	60 (71.4)
Living alone, n (%)
Yes	96 (58.5)	52 (65.0)	44 (52.4)	.102
No	68 (41.5)	28 (35.0)	40 (47.6)	.102
Falls risk factors
Falls in the last 12 months, n (%)
No Falls	77 (47.0)	34 (42.5)	43 (51.2)	.267
1 or more falls	87 (53.1)	46 (57.5)	41 (48.8)	.267
Vision problems, n (%)
Yes	95 (57.9)	44 (55.0)	51 (60.7)	.460
No	69 (42.1)	36 (45.0)	33 (39.3)	.460
Fear of falling, n (%)
Afraid	120 (73.2)	59 (73.8)	61 (72.6)	.871
Not afraid	44 (26.8)	21 (26.3)	23 (27.4)	.871
Use of walking aid, n (%)
Yes	87 (53.1)	60 (75.0)	27 (32.1)	<.001
No	77 (47.0)	20 (25.0)	57 (67.9)	<.001
ABC scale (M ± SD)^a	68.6 ± 21.8	61.8 ± 23.2	74.6 ± 18.6	<.001
Medications (M ± SD)
Total prescription	5.8 ± 4.1	6.8 ± 4.5	4.8 ± 3.4	.002
High-risk prescription	1.5 ± 1.3	1.8 ± 1.5	1.2 ± 1.1	.003
Over-the-counter	2.8 ± 2.3	2.3 ± 2.1	3.2 ± 2.5	.017
VR-12 physical health (M ± SD)^b	39.2 ± 10.6	35.8 ± 9.9	42.5 ± 10.4	<.001
VR-12 mental health (M ± SD)^b	50.7 ± 10.2	48.8 ± 11.7	52.6 ± 8.1	.016
Home risk score (M ± SD)^c	6.7 ± 2.7	6.4 ± 2.8	7.0 ± 2.7	.169

Note. ABC = Activities-specific Balance Confidence; VR-12 = Veterans RAND–12. Percentages are column based.

Only calculated for those individuals who answered nine or more of the ABC questions (n = 152).

VR-12 scores are weighted to a population mean of 50. Higher scores denote better physical or mental health.

Higher scores denote more fall hazards found in the home.

Over half of the participants reported at least one fall in the last 12 months (53%; Table 1). Of the 87 falls, 38 participants (44%) required assistance getting up after the fall, four (5%) resulted in a hip fracture, and 35 (40%) resulted in an injury other than a hip fracture (data not shown). The mean number of falls in the last 12 months for the homebound population was 1.8 (SD = 4.5, range = 0-35), and the mean number of falls in the last 12 months for the non-homebound population was 1.1 (SD = 2.0, range = 0-12). The majority of participants had a fear of falling (73%) and just over half used a walking aid (53%). Non-homebound participants reported better confidence to maintain balance on the ABC scale compared with homebound participants.

Homebound participants had significantly more health-related issues than non-homebound participants, including lower physical and mental health scores on the VR-12. Homebound participants also took more prescription and high-risk prescription medications than non-homebound participants. The most common prescription medication classes used by the homebound group included cardiovascular medications (40%), central nervous system (CNS) active medications (19%), and hormones and synthetic substitutes (8%) (mean number of prescription medications = 6.8, SD = 4.5). The most common prescription medication classes used by the non-homebound group were cardiovascular medications (39%), CNS active medications (16%), and hormones and synthetic substitutes (8%) (mean number of prescription medications = 4.8, SD = 3.4). More non-homebound participants reported taking Vitamin D than homebound participants (p value = .026, data not shown).

Risk Factors for Falling

Increasing age and Black race were protective for falling in both populations (Table 2). However, the association for age was more protective among homebound older adults. Homebound men had more than 3 times the fall rate compared with homebound women (PR = 3.15, 95% CI = [1.11, 8.99]) until examined independently of age, race, and education (PR = 1.94, 95% CI = [0.89, 4.22]).

Table 2.

Association Between Falls Risk Factors and Prevalence of Falling Among Homebound and Non-Homebound Populations.

Risk factor	Homebound		Non-homebound
Risk factor	PR (95% CI)^a	PR (95% CI)^b	PR (95% CI)^c	PR (95% CI)^d
Age (scaled 5 years)	0.70 [0.53, 0.92]	0.70 [0.53, 0.93]	0.88 [0.70, 1.09]	0.88 [0.69, 1.11]
Gender
Male	3.15 [1.11, 8.99]	1.94 [0.89, 4.22]	1.62 [0.70, 3.72]	1.52 [0.73, 3.15]
Female	Reference	Reference	Reference	Reference
Race
Black	0.41 [0.15, 1.08]	0.40 [0.18, 0.87]	0.45 [0.19, 1.07]	0.48 [0.19, 1.24]
White	Reference	Reference	Reference	Reference
Education
<High school	0.83 [0.34, 2.04]	0.88 [0.39, 1.96]	0.65 [0.12, 3.54]	1.34 [0.29, 6.28]
High school	Reference	Reference	Reference	Reference
>High school	1.45 [0.43, 4.88]	0.89 [0.41, 1.91]	0.98 [0.41, 2.35]	0.87 [0.36, 2.15]
Living status
Does not live alone	1.01 [0.35, 2.90]	1.20 [0.58, 2.44]	0.76 [0.35, 1.62]	0.79 [0.37, 1.68]
Lives alone	Reference	Reference	Reference	Reference
Vision problems
Yes	5.50 [2.44, 12.39]	3.77 [2.07, 6.87]	1.84 [0.87, 3.89]	1.79 [0.90, 3.55]
No	Reference	Reference	Reference	Reference
Fear of falling
Afraid	1.49 [0.39, 5.69]	1.45 [0.49, 4.26]	1.36 [0.57, 3.26]	1.45 [0.73, 2.88]
Not afraid	Reference	Reference	Reference	Reference
Use of walking aid
Yes	3.39 [1.42, 8.07]	3.35 [1.61, 6.95]	1.23 [0.60, 2.52]	1.76 [0.87, 3.57]
No	Reference	Reference	Reference	Reference
Medications
Total prescription	1.10 [0.98, 1.24]	1.06 [0.98, 1.16]	1.06 [0.96, 1.18]	1.09 [1.00, 1.18]
No. high-risk prescription	1.53 [1.17, 2.01]	1.36 [1.11, 1.66]	1.29 [0.93, 1.79]	1.31 [1.00, 1.72]
No. over-the-counter	1.15 [0.97, 1.36]	1.16 [1.01, 1.34]	1.04 [0.89, 1.20]	1.02 [0.89, 1.17]
VR-12 mental health^c	0.78 [0.63, 0.96]	0.83 [0.69, 0.99]	0.82 [0.69, 0.98]	0.81 [0.69, 0.95]
VR-12 physical health^c	—	0.92 (0.81, 1.05)	—	0.91 (0.76, 1.08)
ABC scale^d	1.02 [0.93, 1.11]	0.93 [0.88, 0.98]	0.92 [0.83, 1.01]	0.91 [0.85, 0.97]
Home risk score	1.18 [1.05, 1.32]	1.10 [0.97, 1.25]	0.89 [0.81, 0.97]	0.86 [0.78, 0.95]

Note. PR = prevalence ratio; CI = confidence interval; VR-12 = Veterans RAND–12; ABC = Activities-specific Balance Confidence.

Multivariate negative binomial regression with robust variance estimation, adjusted for VR-12 physical health scores.

Multivariate negative binomial regression with robust variance estimation, adjusted for age, race, and education.

Scaled 5 points. Higher scores denote worse health.

Scaled 5 points.

Risk factors for falls observed in both the homebound and non-homebound populations are consistent with what is known in the literature about fall risks in community-dwelling older adults (Deandrea et al., 2010; Nevitt et al., 1989; Tinetti, Speechly, & Ginter, 1988). However, the magnitude of risk was higher in the homebound than in the non-homebound population. This was particularly true for vision impairment (PR = 5.50, 95% CI = [2.44, 12.39] vs. PR =1.84, 95% CI = [0.87, 3.89]), use of a walking aides (PR = 3.39, 95% CI = [1.42, 8.07] vs. PR = 1.23, 95% CI = [0.60, 2.50]), and number of high-risk medications (PR = 1.53, 95% CI = [1.17, 2.01] vs. PR = 1.29, 95% CI = [0.93, 1.79]). The magnitude of risk was also higher for homebound older adults who had a fear of falling and those taking more over-the-counter medications, but these findings did not reach statistical significance.

The presence of home safety hazards affected the rate of falls differently in the two populations. The presence of home hazards was associated with a significantly higher rate of falls among homebound older adults (PR = 1.18, 95% CI = [1.05, 1.32]) and a lower rate of falls among non-homebound older adults (PR = 0.89, 95% CI = [0.81, 0.97]).

Falls Prevention Behaviors and Education

Participants were performing an average of six AGS-recommended falls prevention behaviors. The most common activities performed included exercise (71%), vision check in the last 12 months (74%), use of anti-slip mats in the bathroom (61%), avoiding the use of step stools (65%), and use of safe footwear to avoid falls (91%) (Table 3). Non-homebound older adults were significantly more likely to report exercising on a regular basis compared with homebound older adults (p < .001), whereas homebound older adults were significantly more likely to avoid the use of step stools in their home (p < .001) . There were no significant differences in behaviors between homebound and non-homebound participants for medication reviews and most home safety behaviors.

Table 3.

Falls Prevention Behaviors Among Homebound and Non-Homebound Populations.

Current falls prevention behaviors	TotalN = 164	Homeboundn = 80	Non-homeboundn = 84	p value
Current falls prevention behaviors	n (%)	n (%)	n (%)	p value
Exercise regularly
Yes	116 (70.7)	42 (52.5)	74 (88.1)	<.001
No	48 (29.3)	38 (47.5)	10 (11.9)	<.001
Vision checked in the last 12 months
Yes	120 (74.1)	50 (64.1)	70 (83.3)	.005
No	42 (25.9)	28 (35.9)	14 (16.7)	.005
Medication reviews for falls risk in last 12 months
Yes	46 (28.2)	25 (31.3)	21 (25.3)	.913
No	108 (66.3)	52 (65.0)	56 (67.5)
No medication	9 (5.5)	3 (3.8)	6 (7.2)
Night lights in bedroom and bathroom
Yes	92 (56.1)	42 (52.5)	50 (59.5)	.366
No	72 (43.9)	38 (47.5)	34 (40.5)	.366
Anti-slip mats in bathroom
Yes	100 (61.4)	47 (58.8)	53 (63.9)	.505
No	63 (38.7)	33 (41.2)	30 (36.1)	.505
Grab bars in bathroom
Yes	95 (58.3)	50 (62.5)	45 (54.2)	.285
No	68 (41.7)	30 (37.5)	38 (45.8)	.285
Hand rails on both sides of stairs
Yes	46 (28.1)	23 (28.8)	23 (27.4)	.062
No	58 (35.4)	22 (27.5)	36 (42.9)
No stairs in home	60 (36.6)	35 (43.8)	25 (29.8)
Avoids step stools
Yes	106 (64.6)	63 (78.8)	43 (51.2)	<.001
No	58 (35.4)	17 (21.3)	41 (48.8)	<.001
Use of safe footwear
Yes	148 (91.4)	69 (88.5)	79 (94.1)	.208
No	14 (8.6)	9 (11.5)	5 (6.0)	.208
Throw rugs present and secure
Yes	110 (67.5)	56 (70.9)	54 (64.3)	.308
No	52 (32.5)	22 (29.1)	30 (35.7)	.308
Limited activities due to fear of falling
Yes	74 (45.7)	45 (57.0)	29 (34.9)	0.005
No	88 (54.3)	34 (43.0)	54 (65.1)	0.005

Note. Percentages are column based.

Approximately half of the participants reported that they had limited activities due to a fear of falling (n = 74, 46%). Homebound older adults were significantly more likely to report limited activities compared with non-homebound participants (p = .005). The most commonly reported activities limited due to a fear of falling among homebound older adults included leisure time activities (e.g., visiting with friends), chores inside the home or that required travel into the community, and activities of daily living (e.g., dressing, bathing, and feeding) (data not shown). For non-homebound older adults, participants reported they also limited leisure time activities and chores at home or that required travel. They also commonly reported limiting activities seen as dangerous (e.g., climbing ladders).

Few older adults reported having taken part in a formal falls prevention program (n = 12, 7%), or knew of a falls prevention program being offered near their homes (n = 17, 10%) (Table 4). Although the prevalence was low, more non-homebound older adults participated in a program than homebound older adults (p = .033). Just over one third of the older adults reported being counseled by a health care provider about falls prevention (n = 58, 35%), which was similar in the non-homebound (39%) and homebound (32%) populations. Of those who did receive counseling, the most common advice recalled was to “avoid risk” (52%), modify the home environment (e.g., remove throw rugs; 30%), and use a walker or cane (23%).

Table 4.

Falls Prevention Education Reported by Homebound and Non-Homebound Populations.

Falls prevention education	TotalN = 164	Homeboundn = 80	Non-Homeboundn = 84	p value
Falls prevention education	n (%)	n (%)	n (%)	p value
Participated in falls prevention program
Yes	12 (7.3)	2 (2.5)	10 (11.9)	.033
No	152 (92.7)	78 (97.5)	74 (88.1)	.033
Falls program near home
Yes	17 (10.4)	4 (5.0)	13 (15.5)	.039
No	147 (89.6)	76 (95.0)	71 (84.5)	.039
Counseled by health care professional
Yes	58 (35.4)	31 (38.8)	27 (32.1)	.378
No	106 (64.6)	49 (61.3)	57 (67.9)	.378
Recommendations by health care provider
Modify home environment	17 (30.4)	6 (20.7)	11 (40.7)	—
Exercise	5 (8.9)	2 (6.9)	3 (11.1)
Avoid risk	29 (51.8)	16 (55.2)	13 (48.1)
Careful with medications	3 (5.4)	2 (6.9)	1 (3.7)
Use walking aids	13 (23.2)	7 (24.1)	6 (22.2)
High fracture risk education	5 (8.9)	4 (13.8)	2 (7.4)
Use proper footwear	3 (5.4)	3 (10.3)	1 (3.7)
Other^a	11 (19.6)	2 (6.9)	7 (25.9)

Note. Percentages are column based.

Other includes importance of medical visits and vision checks.

Discussion

Consistent with the existing literature, risk factors for falls among community-dwelling older adults in our study included vision impairment, use of high-risk medications, and use of a walking aid (Deandrea et al., 2010; Nevitt et al., 1989; Tinetti, Speechly, & Ginter, 1988). We found, however, that the magnitude of these risks was greater among homebound participants than non-homebound participants, even after adjusting for physical health.

Most falls prevention programs for frail older adults are delivered in hospital or nursing home settings (Cameron et al., 2012). Falls prevention and intervention research is much needed for frail older adults who are still living in their homes. Having efficacious programs, in addition to an infrastructure that will support long-term implementation, is becoming increasingly important as the population of older adults increases and they are choosing to age in place (Centers for Disease Control and Prevention [CDC], 2013).

Inconsistent with the existing literature, we found that increasing age was protective for falls in both the homebound and non-homebound populations. This is may be due to a selective survival bias, where increasing age disproportionally removed older adults in both populations due to death or having been moved to institutional care because of a fall (Glymour & Greenland, 2008). This may also be due to the cross-sectional nature of the study. Specifically, older adults who fell in the previous 12 months may have reduced their exposure to a fall, and this may have been more likely to occur as the populations aged. We found that men had higher fall rates than women, which is also inconsistent with existing studies. This association was particularly striking among the homebound population until it was examined independently of age, race, and education.

At least two thirds of the older adults in the study were exercising regularly, getting annual vision checks, wearing appropriate non-slip footwear, avoiding step stools, and securing throw rugs (or did not have them). However, only about one third indicated they received falls prevention information from a health care provider, and less than 10% had participated in a falls prevention program. This is consistent with research showing how health care providers perceive falls to be of low importance among other chronic conditions facing older adults and how very few routinely assess falls risks or make referrals to falls prevention programming (Smith et al., 2015). Older adults may be receiving falls prevention information from someone other than a health care professional, such as through media or their senior center, or are engaging in behaviors that reduce falls risk without recognizing it. In either case, capitalizing on current behaviors and motivations for implementation would be important in developing dissemination strategies for evidence-based programs (Gillespie et al., 2012).

Efforts have been made to integrate evidence-based programs into organizations that serve older adults. Uptake of these programs in medical practices, home health organizations, and senior centers has been mixed (Baker, Gottschalk, & Bianco, 2007; Fortinsky et al., 2008; Li et al., 2008; Tinetti et al., 2008). Furthermore, older adult participation in programs offered through these organizations and changes in their falls risk behaviors have been low. Many older adults do not recognize their falls risk, and among those who do, they do not prioritize it as a condition needing intervention (Jansen et al., 2015). Therefore, moving toward dissemination strategies that can incorporate older adults’ existing falls prevention behaviors, whether the behavior is done to reduce falls risk or not, in addition to their perceived risks for falling could be explored in future studies.

Thirty-six percent of homebound participants and 28% of non-homebound participants were getting their medications reviewed for falls risk, and participants were taking an average of 1.48 (SD = 1.35) high-risk medications. Among all recommendations offered in multifactorial falls prevention programs whether targeted for homebound or non-homebound older adults, adherence to recommended medication changes is the poorest (de Vries et al., 2010; Nyman & Victor, 2012). On average, adherence ranges from 0% to less than 50% for medication recommendations within multifactorial programs and for single-component medication interventions. Medication interventions are effective in reducing falls risk (Iyer, Naganathan, McLachlan, & Le Conteur, 2008; van der Velde, Stricker, Pols, & van der Cammen, 2007) but only when providers make medication changes, or approve pharmacist-recommended changes, and older adults adhere to these recommendations. Further falls prevention research should focus on how to improve provider uptake of medication reviews and older adult implementation of recommendations.

This study has several limitations that need to be considered when interpreting the findings. First, this was a cross-sectional study in which participants were asked about current risk factors for falling and about the number of falls they experienced in the previous year. Therefore, temporality is a concern. However, most of our findings are consistent with the existing literature, and those that are not are likely attributed to selection biases. Second, the prevalence of reported falls prevention behaviors may be over-inflated due to social desirability. Third, the homebound and non-homebound study populations were convenience samples, which may limit generalizability of study findings. Finally, there are important chronic health conditions (e.g., depression) that place older adults at risk for a fall (Chang & Do, 2015; Kvelde et al., 2015) but which we did not collect in this study.

We found that homebound status modified the associations between falls risk factors and the prevalence of falls among community-dwelling older adults. Specifically, the prevalence of falling was greater among homebound older adults for most risk factors examined, compared with older adults who were more mobile. Homebound populations can be difficult to reach with falls prevention programming because they often rely on health care and home care services to offer the programs or other caregivers to identify them. Home-based programs have shown great success in reducing falls among older adults (Gillespie et al., 2012). However, the extent to which these programs are available to homebound older adults, and adaptable to their specific needs, is not clear. As we move toward translating evidence-based falls prevention programs into practice for community-dwelling older adults, it will be important to understand how factors influencing access, acceptability, and adherence may differ among those who are homebound and those who can access programs through such community programs and resources as senior centers.

Footnotes

Acknowledgements

We would like to thank all the senior centers and Meals on Wheels programs that allowed us to use their facilities to recruit research participants. In addition, we would like to thank all research staff who helped recruit and conduct the interviews for the study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention (Grant R49 PA-CE07-001).

Author Biographies

Carri Casteel, MPH, PhD, is an associate professor of occupational and environmental health at the University of Iowa (UI) College of Public Health and the associate director for Training and Education for the UI Injury Prevention Research Center. She is an epidemiologist by training with expertise in intervention evaluation and translational research. Her primary research interests are in injury and violence prevention.

Jennifer Jones, MPH, is a doctoral candidate in the Department of Epidemiology at the UNC Gillings School of Public Health and is a graduate research assistant for the Carolina Center for Health Informatics and the UNC Injury Prevention Research Center. Her research interests are in pharmaceutical contributors to older adult falls and injury.

Paula Gildner, MPH, is a program director with the UNC Injury Prevention Research Center (IPRC). She has managed several large-scale, federally funded grants through the IPRC as well as the Carolina Population Center. She is a trained health educator with expertise in reaching hard to reach populations with public health programs.

James M. Bowling, PhD, is a research associate professor of health behavior in the UNC Gillings School of Global Public Health, associate director for Research for the UNC Injury Prevention Research Center, and assistant director for the Carolina Survey Research Laboratory. His expertise is in biostatistics and survey sampling methods.

Susan J. Blalock, PhD, MPH, is a professor in the Division of Pharmaceutical Outcomes and Policy in the UNC Eshelman School of Pharmacy. She is a behavioral scientist with expertise in the area of patient and public health education, and is particularly interested in risk communication and the impact of pharmaceutical care, including patient education and counseling, on patient health outcomes.

References

American Geriatrics Society/British Geriatrics Society. (2011). Summary of the updated American Geriatrics Society/British Geriatrics Society clinical practice guideline for prevention of falls in older persons. Journal of the American Geriatrics Society, 59, 148-157. doi:10.1111/j.1532-5415.2010.03234.x

Baker

D. I.

Gottschalk

Bianco

L. M.

(2007). Step by step: Integrating evidence-based fall-risk management into senior centers. The Gerontologist, 47, 548-554.

Bunn

Dickenson

Barnett-Page

McInnes

Horton

(2008). A systematic review of older people’s perceptions of facilitators and barriers to participation in falls-prevention interventions. Ageing & Society, 28, 449-472. doi:10.1017/S0144686X07006861

Cadore

E. L.

Rodriguez-Manas

Sinclair

Izquierdo

(2013). Effects of different exercise interventions on risk of falls, gait ability, and balance in physically frail older adults: A systematic review. Rejuvenation Research, 16, 105-114. doi:10.1089/rej.2012.1397

Callahan

C. M.

Unverzagt

F. W.

Hui

S. L.

Perkins

A. J.

Hendrie

H. C.

(2002). Six-item screener to identify cognitive impairment among potential subjects for clinical research. Medical Care, 40, 771-781.

Cameron

I. D.

Gillespie

L. D.

Robertson

M. C.

Murray

G. R.

Hill

K. D.

Cumming

R. G.

Kerse

(2012). Interventions for preventing falls in older people in care facilities and hospitals. Cochrane Database of Systematic Reviews, 12, Article CD005465. doi:10.1002/14651858.CD005465.pub3

Centers for Disease Control and Prevention. (2013). The state of aging and health in America. Atlanta, GA: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services.

Chang

V. C.

Do.

M. T

. (2015). Risk factors for falls among seniors: Implications of gender. American Journal of Epidemiology, 181, 521-531. doi:10.1093/aje/kwu268

Danilovich

M. K.

Corcos

D. M.

Marquez

D. X.

Eisenstein

A. R.

Hughes

S. L.

(2015). Performance measures, hours of caregiving assistance, and risk of adverse care outcomes among older adult users of Medicaid home and community-based services. SAGE Open Medicine, 3, Article 2050312115614588. doi:10.1177/2050312115614588.

10.

Deandrea

Lucenteforte

Bravi

Foschi

La Vecchia

Negri

(2010). Risk factors for falls in community-dwelling older people: A systematic review and meta-analysis. Epidemiology, 21, 658-668. doi:10.1097/EDE.0b013e3181e89905.

11.

de Vries

O. J.

Peeters

G. M. E. E.

Elders

P. J. M.

Muller

Knol

D. L.

Danner

S. A.

. . . Lips

. (2010). Multifactorial intervention to reduce falls in older people at high risk of recurrent falls: A randomized controlled trial. Archives of Internal Medicine, 170, 1110-1117. doi:10.1001/ archinternmed.2010.169

12.

Faes

M. C.

Reelick

M. F.

Melis

R. J.

Borm

G. F.

Esselink

R. A.

Rickkert

M. G. O.

(2011). Multifactorial fall prevention for pairs of frail community-dwelling older fallers and their informal caregivers: A dead end for complex interventions in the frailest fallers. Journal of the American Medical Directors Association, 12, 451-458. doi:10.1016/j.jamda.2010.11.006

13.

Fortinsky

R. H.

Baker

D. I.

Gottschalk

King

Trella

Tinetti

M. E.

(2008). Extent of implementation of evidence-based fall prevention practices for older patients in home health care. Journal of the American Geriatrics Society, 56, 737-743. doi:10.1111/j.1532-5415.2007.01630.x

14.

Ganz

D. A.

Bao

Shekelle

P. G.

Rubenstein

L. Z.

(2007). Will my patient fall? Journal of American Medical Association, 297, 77-86.

15.

Gillespie

L. D.

Robertson

M. C.

Gillespie

W. J.

Sherrington

Gates

Clemson

L. M.

Lamb

S. E.

(2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews, 9, Article CD007146. doi:10.1002/14651858.CD007146.pub3

16.

Glymour

M. M.

Greenland

(2008). Causal diagrams. In Rothman

K. J.

Greenland

Lash

T. L.

(Eds.), Modern epidemiology (pp. 183-212). Philadelphia, PA: Lippincott William & Wilkins.

17.

Hartikainen

Lönnroos

Louhivuori

(2007). Medication as a risk factor for falls: Critical systematic review. The Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 62, 1172-1181.

18.

Hendriks

M. R. C.

Bleijlevens

M. H. C.

van Haastregt

J. C.

Crebolder

H. F.

Diederiks

J. P. M.

Evers

S. M. A. A.

. . . van Eijk

J. T.

(2008). Lack of effectiveness of a multidisciplinary fall-prevention program in elderly people at risk: A randomized, controlled trial. Journal of the American Geriatrics Society, 56, 1390-1397. doi:10.1111/j.1532-5415.2008.01803.x

19.

Herr

Latouche

Ankri

(2013). Homebound status increases death risk within two years in the elderly: Results from a national longitudinal survey. Archives of Gerontology and Geriatrics, 56, 258-264. doi:10.1016/j.archger.2012.10.006

20.

Hickam

D. H.

Weiss

J. W.

Guise

J. M.

Buckley

Motu’apuaka

Graham

. . . Saha

(2013). Outpatient case management for adults with medical illness and complex care needs (Comparative Effectiveness Review No. 99, AHRQ Publication No.13-EHC031-EF). Rockville, MD: Agency for Healthcare Research and Quality.

21.

Houston

D. K. S.

Tooze

J. A.

Demons

J. L.

Davis

B. L.

Shertzer-Skinner

Kearsley

L. B.

. . . Williamson

J. D.

(2015). Delivery of a vitamin D intervention in homebound older adults using a Meals-on-Wheels program: A pilot study. Journal of the American Geriatrics Society, 63, 1861-1867. doi:10.1111/jgs.13610

22.

Iyer

Naganathan

McLachlan

A. J.

Le Conteur

D. G.

(2008). Medication withdrawal trials in people aged 65 years and older. Drugs & Aging, 25, 1021-1031. doi:10.2165/0002512-200825120-00004

23.

Jansen

Schoe

van Rijn

Abu-Hanna

Moll van Charante

E. P.

van der Velde

de Rooij

S. E.

(2015). Factors associated with recognition and prioritization for falling, and the effect on fall incidence in community dwelling older adults. BMC Geriatrics, 15, Article 169. doi:10.1186/s12877-015-0165-2

24.

Kazis

L. E.

Miller

D. R.

Clark

J. A.

Skinner

K. M.

Lee

Ren

X. S.

. . . Ware

J. E.

(2004). Improving the response choices on the veterans SF-36 health survey role functioning scales: Results from the Veterans Health Study. Journal of Ambulatory Care Management, 27, 263-280.

25.

Kvelde

Lord

S. R.

C. T.

Reppermind

Kochan

N. A.

Sachdev

. . . Delbaere

(2015). Depressive symptoms increase fall risk in older people, independent of antidepressant use, and reduced executive and physical functioning. Archives of Gerontology and Geriatrics, 60, 190-195. doi:10.1016/j.archger.2014.09.003

26.

Lajoie

Gallagher

S. P.

(2004). Predicting falls within the elderly community: Comparison of postural sway, reaction time, the Berg balance scale and the Activities-specific Balance Confidence (ABC) scale for comparing fallers and non-fallers. Archives of Gerontology and Geriatrics, 38, 11-26.

27.

Levine

S. A.

Boal

Boling

P. A.

(2003). Home care. Journal of American Medical Association, 290, 1203-1207.

28.

Harmer

Glasgow

Mack

K. A.

Sleet

Fisher

K. J.

. . . Tompkins

(2008). Translation of an effective Tai Chi intervention into a community-based falls-prevention program. American Journal of Public Health, 98, 1195-1198. doi:10.2105/AJPH.2007.120402

29.

McMahon

Talley

K. M.

Wyman

J. F.

(2011). Older people’s perspectives on fall risk and fall prevention programs: A literature review. International Journal of Older People Nursing, 6, 289-298. doi:10.1111/j.1748-3743.2011.00299.x

30.

National Center for Injury Prevention and Control. (2005). Check for safety: A home fall prevention checklist for older adults. Retrieved from http://www.cdc.gov/homeandrecreationalsafety/pubs/english/booklet_eng_desktop-a.pdf

31.

National Center for Injury Prevention and Control. (2014). WISQARS fatal and nonfatal injury reporting system, 2014. Retrieved from http://www.cdc.gov/injury/wisqars

32.

Nevitt

M. C.

Cummings

S. R.

Kidd

Black

(1989). Risk factors for recurrent nonsyncopal falls: A prospective study. Journal of American Medical Association, 261, 2663-2668.

33.

Nyman

S. R.

Victor

C. R.

(2012). Older people’s participation in and engagement with falls prevention interventions in community settings: An augment to the Cochrane systematic review. Age and Ageing, 41, 16-23. doi:10.1093/ageing/afr103

34.

Powell

L. E.

Myers

A. M.

(1995). The Activities-specific Balance Confidence (ABC) scale. The Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 50A, M28-M34.

35.

Qui

W. Q.

Dean

Liu

George

Gann

Cohen

Bruce

M. L.

(2010). Physical and mental health of the homebound elderly: An overlooked population. Journal of the American Geriatrics Society, 58, 2423-2428. doi:10.1111/j.1532-5415.2010.03161.x

36.

Rubenstein

L. Z.

Josephson

K. R.

(2006). Falls and their prevention in elderly people: What does the evidence show? Medical Clinics of North America, 90, 807-824.

37.

Scheffer

A. C.

Schuurmans

M. J.

van Dijk

van der Hooft

deRooij

S. E.

(2008). Fear of falling: Measurement strategy, prevalence, risk factors and consequences among older persons. Age and Aging, 37, 19-24. doi:10.1093/ageing/afm169

38.

Smith

M. L.

Stevens

J. A.

Ehrenreich

Wilson

A. D.

Schuster

R. J.

Cherry

C. O.

Ory

M. G.

(2015). Healthcare providers’ perceptions and self-reported fall prevention practices: Findings from a large New York health system. Frontiers in Public Health, 3, Article 17. doi:10.3389/fpubh.2015.00017.

39.

Stenhagen

Ekström

Nordell

Elmståhl

(2013). Falls in the general elderly population: A 3- and 6-year prospective study of risk factors using data from the longitudinal population study “Good ageing in Skane.” BMC Geriatrics, 13, 1-11. doi:10.1186/1471-2318-13-81

40.

Stevens

J. A.

Mack

K. A.

Paulozzi

L. J.

Ballesteros

M. F.

(2008). Self-reported falls and fall-related injuries among persons aged at least 65 years—United States, 2006. Journal of Safety Research, 39, 345-349. doi:10.1016/j.jsr.2008.05.002

41.

Stevens

J. A.

Noonan

R. K.

Rubenstein

L. Z.

(2010). Older adult fall prevention: Perceptions, beliefs, and behaviors. American Journal of Lifestyle Medicine, 4, 16-20. doi:10.1177/1559827609348350

42.

Tinetti

M. E.

Baker

D. I.

King

Gottschalk

Murphy

T. E.

Acampora

. . . Allore

H. G.

(2008). Effect of dissemination of evidence in reducing injuries from falls. New England Journal of Medicine, 359, 252-261. doi:10.1056/NEJMoa0801748

43.

Tinetti

M. E.

Speechley

Ginter

S. F.

(1988). Risk factors for falls among elderly persons living in the community. New England Journal of Medicine, 319, 1701-1707

44.

van der Velde

Stricker

B. H. C.

Pols

H. A.

van der Cammen

T. J

. (2007). Risk of falls after withdrawal of fall-risk-increasing drugs: A prospective cohort study. British Journal of Clinical Pharmacology, 63, 232-237.

45.

Woolcott

J. C.

Richardson

K. J.

Wiens

M. O.

Patel

Marin

Khan

K. M.

Marra

C. A.

(2009). Meta-analysis of the impact of 9 medication classes on falls in elderly persons. Archives of Internal Medicine, 169, 1952-1960. doi:10.1001/archinternmed.2009.357.