The purpose of this study was to examine the prevalence of falls and risk factors for falls in 1,515 adults (≥ 18 years) with intellectual disability using baseline data from the Longitudinal Health and Intellectual Disability Study. Nearly 25% of adults from the study were reported to have had one or more falls in the past 12 months. The prevalence of falls increased with advancing age. A series of univariate and multivariate logistic regressions were performed to identify risk factors for falls in the full sample and in subsamples. The risk factors for falls in adults with intellectual disability are being female, having arthritis, having a seizure disorder, taking more than 4 medications, using walking aids, and having difficulty lifting/carrying greater than 10 lb.
Falls are a major public health concern, affecting approximately one third of individuals aged 65 years or older each year (Fisher & Kettl, 2005; Hausdorff, Rios, & Edelberg, 2001; Stevens, Mack, Paulozzi, & Ballesteros, 2008), but are a particularly troublesome issue for adults with intellectual disability. Studies indicate that the prevalence rate of falls among adults with intellectual disability living in community settings is substantially higher than in the general population. In several recent studies, the prevalence of falls among adults with intellectual disability (20 to 80 years) ranged from 29% to 57% (Chiba et al., 2009; Cox, Clemson, Stancliffe, Durvasula, & Sherrington, 2010; Finlayson, Morrison, Jackson, Mantry, & Cooper, 2010; Wagemans & Cluitmans, 2006). Among adults living in congregate care settings (e.g., institutions, group homes), the rate of falls was even higher: 70% of residents aged from 18 to 77 years fell in a 5-year period, and 79% of those falls resulted in injury (Grant, Pickett, Lam, O'Connor, & Ouellette-Kuntz, 2001). Yet, prevalence rates of falls for adults with intellectual disability across age groups have not been well established.
Falls are a major cause of serious injuries and hospitalization for adults with intellectual disability, with approximately 50% to 61% of injuries requiring medical care (Hsieh, Heller, & Miller, 2001; Sherrard, Tonge, & Ozanne-Smith, 2001; Wang, McDermott, & Sease, 2002). Fractures are estimated to occur 1.7 to 3.5 times more frequently in people with intellectual disability than their counterparts in the general population (Lohiya, Crinella, Tan-Figueroa, Caires, & Lohiya, 1999; Sherrard et al., 2001; Tannenbaum, Lipworth, & Baker, 1989). As a result, individuals who sustain a fall-related injury often require more assistance with daily living and have less mobility, which could lead to precocious nursing home placement and further health risks (e.g., pressure ulcers, deconditioning). In addition, falls may also result in psychosocial changes such as increased fear of falling, causing individuals to reduce community activity and increase their risk of social isolation.
Risk Factors for Falls in the General Population
A social epidemiology, conceptual model proposed by Cwikel and Fried (1992) offers a useful multifactoral approach to organize the risk factors for falls in the general population. This model includes demographic, medical, and psychosocial host factors and environmental factors.
Research on risks for falls in the general population has identified older age (Rubenstein, 2006; Samelson, Zhang, Kiel, Hannan, & Felson, 2002; Stevens et al., 2008) and being female (Williams, Kool, Robinson, & Ameratunga, 2012) as the key demographic factors. Identification of risk factors for falls is crucial for developing fall prevention guidelines and recommendations for prevention of falls in adults with intellectual disability. Research in risk factors for falls among adults with intellectual disability is quite limited. The few studies that examined risks for falls among adults with intellectual disability found that the demographic risks included older age/increasing age (Cox et al., 2010; Hsieh et al., 2001), being male and having a mild level of intellectual disability (Grant et al., 2001), and having a diagnosis of cerebral palsy (Cox et al., 2010), whereas the presence of Down syndrome is a protector (Finlayson et al., 2010).
Key medical, identified risk factors in the general population include specific health conditions (e.g., postural hypotension, osteoporosis, arthritis, epilepsy, stroke), sensory and perceptual declines (e.g., visual deficit, gait, and balance deficit), symptoms and acute diseases (e.g., dizziness, cognitive impairment), functional limitations (e.g., mobility limitation, use of assistive device, reduced lower muscle strength, urinary incontinence), use of medications (e.g., polypharmacy, hypnosedatives, antidepressants, and antipsychotics), and a history of falls (American Geriatrics Society, British Geriatrics Society, & American Academy of Orthopaedic Surgeons Panel on Falls Prevention, 2001; Ganz, Bao, Shekelle, & Rubenstein, 2007; Hartikainen, Lönnroos, & Louhivuori, 2007; Rubenstein & Josephson, 2006; Woolcott et al., 2009).
For individuals with intellectual disability, key medical risk factors reported are seizure disorder (Hsieh et al., 2001; Sherrard, Tonge, & Ozanne-Smith, 2002; Wagemans & Cluitmans, 2006), urinary incontinence (Finlayson et al., 2010), visual deficits and past fracture (Cox et al., 2010), abnormal gait pattern (Hale, Bray, & Littmann, 2007), and ambulatory status (Hsieh et al., 2001; Wagemans & Cluitmans, 2006). Although having a seizure disorder has been found to be a consistent factor that contributes to falls among adults with developmental disabilities, it might overshadow other factors for those who do not have seizure disorders. To better understand other factors that are associated with falls for adults with intellectual disability who do not have seizure disorders, a separate analysis is needed to identify the factors.
Because of a higher prevalence of health conditions and preexisting conditions, people with intellectual disability also tend to be on multiple medications (Lee et al., 2011). Recently, a systematic review and a meta-analysis study had a consistent conclusion that the medications associated with an increased risk of falling were benzodiazepines (hypnosedatives), antidepressants, and antipsychotics; a weak association was found between anticonvulsants, antihypertensives, and an increased risk of falling (Hartikainen et al., 2007; Woolcott et al., 2009). These risk factors were not found in studies involving adults with intellectual disability.
In regard to psychosocial factors, depression can be both a risk factor and an outcome of falls, especially for those who suffer devastating injury (Biderman, Cwikel, Fried, & Galinsky, 2002). Yet this factor has not been explored for individuals with intellectual disability.
Environmental factors related to falls include poor lighting, loose carpets, and lack of bathroom safety equipment (Rubenstein & Josephson, 2002). Environmental assessments are not included in the present study, therefore environmental factors are not discussed.
Physical Activity Factors
Although physical inactivity is not listed in this model, it is another risk factor that can increase risk of falls (Mertz, Lee, Sui, Powell, & Blair, 2010). People with intellectual disability tend to have higher levels of physical inactivity and poor physical fitness (Draheim, Williams, & McCubbin, 2002; Peterson, Janz, & Lowe, 2008), poorer dietary habits (e.g., reduced calcium intake, higher fat consumption; McGuire, Daly, & Smyth, 2007), and vitamin D deficiency (Vanlint & Nugent, 2006) compared to the general population. These factors may be associated with reduced muscle mass, bone mineral density (Takata & Yasui, 2001; i.e., osteopenia or osteoporosis), and increased body fat (Roubenoff, 2000), thereby increasing their risk of falls. Several meta-analyses have shown that taking vitamin D supplements benefits fall prevention (Bischoff-Ferrari et al., 2004; Bischoff-Ferrari et al., 2009). Therefore, the level of physical activity and use of vitamin D supplements need to be taken into consideration while examining risk factors for falls in adults with intellectual disability. The present study examines the demographic, medical, and psychosocial factors listed in this model along with the physical activity factors.
Most fall-related studies in the intellectual disability population were with a small sample and limited to residential facilities or clinic settings. To better understand the risk factors associated with falls in adults with intellectual disability across different settings, more large-scale epidemiological studies are needed. The purpose of this retrospective large-scale study was to (a) examine the prevalence of falls among adults with intellectual disability living in community settings; (b) determine if adults with intellectual disability who experienced one or more falls in the past 12 months (termed here fallers) differ in characteristics compared to those who have not fallen (non-fallers); (c) identify the risk factors (demographic, medical, psychosocial, and physical activity factors) associated with falls in adults with intellectual disability after controlling for age, sex, level of intellectual disability, diagnosis of Down syndrome, and cerebral palsy; and (d) identify the risk factors (demographic, medical, psychosocial, and physical activity factors) associated with falls requiring medical care among the fallers after controlling for age, sex, level of intellectual disability, diagnosis of Down syndrome, and cerebral palsy.
Our study used the Longitudinal Health and Intellectual Disability Study baseline data (N = 1,515) to examine the prevalence of falls and explore potential risk factors for falls in adults with intellectual disability. The study sample consisted of 680 females and 835 males with intellectual disability, aged 18 years and older, living in various residential settings across 50 states. Participants ranged in age from 18 to 86 years (M = 37.60, SD = 14.36). Eighty-nine percent of the participants were Caucasian, 6% African American, and 3% Hispanic. Twenty-five percent of participants had Down syndrome; 12% had cerebral palsy; and 13% had autism or pervasive developmental disorder. Over half (55%) of the participants were living with a family member or relative, 13% on their own, 5% with a host family or in foster care home, 15% in supporting living, and the remainder (12%) in small or large supervised residences. Less than two thirds (61%) of the participants were employed. Surveys were completed by parents (46.1%); followed by health care providers (21.8%); residential, day program, or social service staff (13.6%); relatives (8%); nonrelative live-in caregivers (3.6%); and other (2.5%). Less than 5% of the respondents did not report their relationship with adults with intellectual disability.
Longitudinal Health and Intellectual Disability Study
This ongoing 4-year longitudinal study uses the Longitudinal Health and Intellectual Disability Survey (LHIDS) to collect data and to examine the relationships among modifiable health behaviors, risk factors, chronic health conditions, psychosocial well-being, and social participation of adults with intellectual disability. Data for the present study were collected as part of its baseline data (N = 1,615). After excluding missing data on falls, data from 1,515 participants were included for analyses. Data were collected from caregivers of adults with intellectual disability between February 2010 and February 2011.
The study recruited family members or primary caregivers of adults with intellectual disability by asking families at Special Olympics events, posting recruitment information in various avenues (e.g., Facebook, newsletter advisements, and recruiting materials distributed at conferences), collaborating with managed care organizations, and communicating about the study with various service agencies. To broaden the participants' views, a mixed-method (mail and online surveys) data collection procedure was used. Informants were asked to complete a paper or online survey, depending on their preference. A total of 2,841 surveys were distributed (2,182 paper and 659 online), and 1,615 surveys were completed and returned (1,179 paper and 436 online) with an overall response rate of 57%. The response rate for the paper surveys was 54%, compared to 66% for the online survey.
The LHIDS survey was developed in three phases: pre-survey preparation, pilot testing, and survey administration. Each of the phases is described below.
Phase I: Pre-survey preparation
A comprehensive review of existing health survey instruments was conducted (e.g., National Health and Nutrition Examination Survey, National Health Interview Survey, Behavioral Risk Factor Surveillance Survey, University of Michigan Health Risk Appraisal Questionnaire, etc.) to determine the most appropriate questions for our cohort. After reviewing the surveys, a list of questions was assembled and reviewed by the project team. Steps were then taken to revise/refine survey questions to make them more appropriate for adults with intellectual disability and their respondent (e.g., family member, health care provider). After completing a draft list of questions, an expert panel reviewed them for content, clarity, and significance to the target population, and provided feedback on each item. After revising the instrument, the Survey Research Laboratory (SRL) at the University of Illinois at Chicago conducted an extensive analysis of the survey items to improve wording, clarity, and design. The survey was then refined several more times with feedback from the SRL and expert review panel. Afterward, the survey was given to 10 informants who cared for an adult with intellectual disability to ensure that questions were comprehensible and appropriate to their son or daughter.
Phase II: Pilot testing
The paper–pencil survey was pilot tested for usability and item clarity with a sample of 10 informants. Volunteers were recruited through a research center's advocacy advisory group (parents, siblings, and self-advocates) and staff in service agencies. Test–retest reliability was conducted with 15 primary caregivers and 15 direct support staff, with test–retest intervals ranging from 2 weeks to 6 weeks. Approximately 52% of the retests were conducted within 2 weeks, 21% within 2–3 weeks, and 27% within 3–6 weeks. For the categorical questions, test–retest reliability (k-statistic) ranged from 0.68 to 0.95, and for the interval questions, the test–retest reliability (intraclass correlation coefficient) ranged from 0.75 to 0.94.
Phase III: Survey administration
A unique computer-generated identification number was placed on each survey included in the survey packet sent to each primary family caregiver or direct care staff. The survey packet consisted of the survey, a letter explaining the purpose of the study, and a participant information sheet. The paper survey took 20–30 min to complete by hand and 15–20 min to complete online.
The LHIDS was divided into three sections: Health and Function, Health Behaviors, and Sociodemographics. Health and Function questions were divided into four subsections:
health status, which included informant-rated Health-related Quality of Life (HRQOL), which was adopted from Centers for Disease Control and Prevention's Core Healthy Days Measures to assess a person's well-being through four questions on informant-rated health, number of recent days when physical health was not good, number of recent days when mental health was not good, and number of recent activity limitation days because of poor physical or mental health (Terry et al., 2005);
36 chronic health conditions, which listed eight categories: (a) bone, joint, and muscles (i.e., arthritis, osteoporosis); (b) heart and circulation (e.g., heart condition, high blood pressure, low blood pressure, etc.); (c) hormonal (i.e., diabetes, thyroid); (d) lungs/breathing (i.e., asthma, chronic bronchitis, sleep disorder); (e) mental health (e.g., anxiety disorder, bipolar disorder, depression, schizophrenia, etc.); (f) pain/discomfort (back pain, constipation, headache, etc.); (g) sensory (dizziness, hearing impairment, vision impairment); and (h) other (cancer, chronic fatigue, seizure disorder, and urinary incontinence); and whether or not medication was being taken for the condition;
physical function (e.g., use of a walking aid, performing physical tasks, falls); and
health care access.
Health Behaviors included five domains:
physical activity/exercise including moderate, vigorous, strength physical activity, Special Olympics participation and training, and sedentary behaviors (i.e., TV viewing time);
weight control and dietary habits;
smoking and alcohol;
oral hygiene; and
Sociodemographics was assessed by recording how often adults with intellectual disability engaged in five community activities within the last month, adopted from the Community and Social Participation instrument developed by Heller, Hsieh, and Rimmer (2004).
The outcome measure in the present study is occurrence of falls. The independent variables are demographic, medical, psychosocial, and physical activity factors.
Occurrence of falls
We asked informants, “How many falls has he/she experienced in the past 12 months?” A fall was defined as “a sudden unintentional change in position causing an individual to land at a lower level, on an object, the floor, or the ground, other than as a consequence of a sudden onset of paralysis, epileptic seizure, or overwhelming external force” (Tinetti, Baker, Dutcher, Vincent, & Rozett, 1997). The responses included 0, 1, 2, 3, and 4 or more. Participants who experienced one or more falls in the past 12 months were grouped as fallers, and those who did not have a fall were defined as non-fallers. Whether the fall required medical care was also asked.
Demographic information included age, sex, race, diagnosis related to intellectual disability (Down syndrome, cerebral palsy, autism), level of intellectual disability, type of residence (least, moderately, and most supported), and employment status.
Chronic health conditions
We asked informants, “Does he/she have any of the following conditions based on a diagnosis from a doctor?” with a list of 36 health conditions. Informants were also asked whether the adults with intellectual disability took medication for each health condition. Chronic health conditions that were included in analysis as independent variables were arthritis, osteoporosis, heart condition, high blood pressure, low blood pressure, stroke, diabetes, sleep disorder, dizziness/vertigo, dementia/Alzheimer's disease, back pain, seizure disorder, hearing impairment, and vision impairment. Obesity was also included as a chronic health condition. Body weight and height were reported by informants. We calculated body mass index (BMI) using the formula
Obesity was defined when BMI was greater than 30 kg/m2.
Functional limitations were assessed by urinary incontinence, use of walking aids, ability to lift/carry greater than 10 lb, climbing a flight of stairs, and walking 3 blocks.
Medication use/vitamin D supplement
Polypharmacy and use of medications associated with falls were included to assess medication use. Polypharmacy was defined as taking four or more prescribed medications. Use of medications included antipsychotics, antidepressants, and hypnosedatives. Taking vitamin D supplements was assessed by asking whether the adult with intellectual disability regularly took vitamin D nutritional supplements.
Psychosocial factors included anxiety disorder and depression.
Physical activity factors
Physical activity. Physical activity was measured by asking, “On average, how many days a week does he/she do moderate physical activities for at least 30 minutes?” The definition of moderate physical activity and examples of moderate physical activities were provided. It was rated from 1 (never) to 4 (4 or more times a week).
Special Olympics participation
We assessed individuals' Special Olympics participation by asking, “Does he/she currently participate in the Special Olympics?”
Descriptive statistics were used to examine the prevalence of falls and frequency of medical care required for fallers. We conducted t-tests to examine whether there were group (fallers vs. non-fallers) differences among the continuous variables (age). For the categorical demographic variables (age group, sex, race, obesity, intellectual disability–related diagnosis, level of intellectual disability, use of walking aids, living arrangement, employment, and Special Olympics participation), chi-square tests were employed to detect group differences. A significance level at a p value of .05 was used for all the analyses. Age, sex, level of intellectual disability, diagnosis of Down syndrome, and cerebral palsy were identified as variables to be controlled in the multivariate logistic regression models.
The purposeful selection process was employed (Hosmer & Lemeshow, 2000). It included a series of univariate binary logistic regressions to identify potential risk factors for falls in the full sample (N = 1,515) and the subsample (n = 1,225) excluding those who had seizure disorders. The dependent variable was occurrence of falls in the past 12 months. The independent variables that included participant demographic factors (age, sex, Down syndrome, cerebral palsy, obesity), medical factors including chronic health conditions (arthritis, heart condition, high blood pressure, low blood pressure, osteoporosis, history of stroke, diabetes, sleep disorder, anxiety disorder, dementia, back pain, dizziness, vision impairment, hearing impairment, seizure disorder, and obesity), functional limitations (urinary incontinence, use of walking aids, difficulty lifting/carrying, climbing stairs, walking), use of medications/vitamin D supplement (polypharmacy and use of medications associated with falls), and physical activity factors (moderate physical activity and Special Olympics participation) were entered separately into a univariate logistic model. Any variables having a significant univariate test at p value cut-off point of .05, based on the Wald test, were selected as candidates for the multivariate analysis. Pearson correlation tests among independent variables were performed to examine collinearity. Only one variable was selected for the multivariate analysis, if a correlation coefficiency was greater than 0.7 between two independent variables. After risk factor candidates were identified, these variables were entered into multivariate logistic regressions while controlling for age, sex, level of intellectual disability, diagnosis of Down syndrome, and cerebral palsy. In the iterative process of variable selection, variables were removed from the model if they were not significant and not a control variable. Significance was evaluated at the 0.1 alpha level and confounding as a change in any remaining parameter estimate greater than 15% as compared to the full model to obtain the final model. Hosmer-Lemeshow Goodness of Fit tests were used at each step to select the best fit final model. (The statistical package used was SPSS for Windows, version 17.0).
Prevalence of Falls
Of the 1,515 participants, 372 (24.6%) were reported to have experienced falls in the past 12 months. The prevalence of falls by age groups was 20.6% between 18–44 years, 32.3% between 45–64 years, and 43.1% for 65 years and older. The overall prevalence of falls for males was 20.0% and 30.1% for females. While taking both age groups and sex into account, the prevalence of falls was 17.0% for males and 25.4% for females between 18–44 years, 27.0% vs. 37.7% between 45–64 years, 35.3% vs. 51.6% for 65 years and older. Within the fallers, 41.7% experienced one fall; one fourth of the fallers had 4 falls or more; 20.2% had 2 falls; and 13.2% had 3 falls. Over half (58%) of the fallers experienced recurrent falls in the past 12 months. Twenty-four percent of the fallers required medical care.
Comparisons Between Fallers and Non-fallers
Table 1 presents the results of the descriptive analyses between fallers and non-fallers. The mean age of the fallers (n = 372) was 41.34 years compared to 36.14 years for the non-fallers (n = 1508), t(1506) = 6.08, p < .001. Over one half of fallers vs. non-fallers (57.8% vs. 73.0%) were in the 18- to 44-year-old age group; 37.4% of fallers (vs. 23.8% non-fallers) were in the 45- to 64-year-old group; and 7.5% of fallers (vs. 3.3% non-fallers) were in the age 65 years and older group, χ2(2) = 33.8, p < .001. Significant differences (p < .05) were found between fallers and non-fallers. Compared with non-fallers, fallers were more likely to be female (55.1% vs. 44.9%), have a diagnosis of cerebral palsy (20.7% vs. 9.7%), have severe intellectual disability (2.58 vs. 2.34), and use a walking aid (23.6% vs. 3.7%), but less likely to have Down syndrome (19.8% vs. 27.2%) or autism (8.2% vs. 14.3%), be employed (50.3% vs. 64%), and participate in Special Olympics (40.5% vs. 48.0%).
Identification of Risk Factors
Risk factors for falls in full sample
Univariate binary logistic regressions were employed in the full sample to determine potential risk factors for falls. The following factors were significant at p < .05: the demographic factors were age, sex, level of intellectual disability, cerebral palsy, and Down syndrome; the medical factors were chronic health conditions (having a condition of arthritis, osteoporosis, stroke, diabetes, sleep disorder, dementia, back pain, dizziness/vertigo, seizure disorder, and vision impairment), functional limitations (urinary incontinence, use of walking aids, difficulty lifting or carrying over 10 lb, difficulty climbing one flight of stairs, difficulty walking three blocks), use of medications/supplement (taking vitamin D supplement, taking antidepressant, hypnosedatives, antipsychotics, and polypharmacy); the psychosocial factors were depression and anxiety disorder; and the physical activity factors were number of days engaging in moderate physical activity and Special Olympics participation.
The significant potential risk factor variables were entered into a multivariate logistic regression using a purposeful selection method, while controlling for age, gender, level of intellectual disability, Down syndrome, and cerebral palsy. The final model is presented in Table 2. The overall final model was significant, χ2(15) = 183.56, p < .001, Nagelkerke R2 = 0.24. The results indicated that being female, OR = 1.44, 95% CI [1.05, 1.97]; having arthritis, OR = 1.75, 95% CI [1.07, 2.87]; having a seizure disorder, OR = 1.94, 95% CI [1.13, 2.41]; having polypharmacy, OR = 1.78, 95% CI [1.18, 2.68]; using walking aids, OR = 2.89, 95% CI [1.61, 5.18]; and having difficulty lifting or carrying over 10 lb, OR = 1.94, 95% CI [1.23, 3.07], were risk factors for falls.
Risk factors for falls for individuals without seizure disorders
A series of univariate binary logistic regressions were performed to determine potential risk factors for falls while excluding those who had a seizure disorder (n = 1,225). The following factors were significant at p < .05: the demographic factors were age, sex, level of intellectual disability, cerebral palsy; the medical factors were chronic health conditions (having a condition of arthritis, osteoporosis, stroke, diabetes, heart condition, dementia, back pain, dizziness/vertigo, seizure disorder, hearing, and vision impairment), functional limitations (urinary incontinence, use of walking aids, difficulty lifting or carrying over 10 lb, difficulty climbing one flight of stairs, difficulty walking three blocks), use of medications/supplements (taking antidepressant and polypharmacy); the psychosocial factors were depression and anxiety disorder; and the physical activity factor was number of days engaging in moderate physical activity.
Table 3 presents the final model of a multiple logistic regression (n = 1,225), while controlling for age, gender, level of intellectual disability, Down syndrome, and cerebral palsy. The overall model was significant, χ2(11) = 135.73, p < .001, Nagelkerke R2 = 0.14. The results indicated that having a higher level of intellectual disability, OR = 1.30, 95% CI [1.07, 1.58]; having arthritis, OR = 1.83, 95% CI [1.08, 3.10]; a heart condition, OR = 1.97, 95% CI [1.16, 3.34]; back pain, OR = 2.01, 95% CI [1.15, 3.51]; and urinary incontinence, OR = 2.06, 95% CI [1.15, 3.51]; using a walking aid, OR = 3.00, 95% CI [1.48, 6.10]; and having difficulty walking 3 blocks, OR = 2.12, 95% CI [1.35, 3.34] were risk factors for falls in adults with intellectual disability who did not have a seizure disorder.
Risk factors for fall-related injuries requiring medical care
A series of univariate binary logistic regressions were performed to determine potential risk factors for falls that required medical attention (n = 372). The following factors were significant at p < .05: the demographic factor was age; the medical factors were chronic health conditions (having osterporosis, heart condition, and back pain), functional limitations (urinary incontinence, using walking aids, and having difficulty walking 3 blocks), and use of medication (taking antidepressant, antipsychotropics, and polypharamcy); the psychosocial factor was depression; and the physical activity factor was number of days engaging moderate physical activities.
Table 4 presents the final multiple logistic regression model, while controlling for age, sex, level of intellectual disability, Down syndrome, and cerebral palsy. The overall model was significant, χ2(8) = 22.52, p < .01, Nagelkerke R2 = 0.12. The results indicated that having osteoporosis, OR = 2.76, 95% CI [1.20, 6.37]; back pain, OR = 2.40, 95% CI [1.14, 5.09]; and urinary incontinence, OR = 2.25, 95% CI [1.08, 4.66]; were risk factors for falls that required medical care.
The findings from this retrospective study show that adults with intellectual disability have a comparable rate of falls at younger ages compared to norm-referenced data in the older ages. One in three middle-aged adults (aged 45–64 years) with intellectual disability living in community settings experienced one or more falls in the past 12 months, which is similar to the rate found in the general older population (Fisher & Kettl, 2005; Hausdorff et al., 2001; Stevens et al., 2008). One in the five young adults (aged 18–44 years) with intellectual disability in community settings experienced one or more falls, but because the rate of falls is presumed to be low in younger populations without disabilities, there are no norm-referenced data to compare our sample to in terms of prevalence of falls in this younger cohort. We also found that females had a higher prevalence of falls compared to males across all age groups and that the prevalence of falls increased with advancing age.
The temporal relationship between occurrence of falls and underlying risk factors for adults with intellectual disability are still emerging. Our findings suggest that falls for adults with intellectual disability were associated with having arthritis, seizure disorder, and polypharmacy, using a walking aid, and having difficulty lifting/carrying greater than 10 lb. Our finding that seizure disorders increased the risk of falls supports previous studies (Finlayson et al., 2010; Hsieh et al., 2001; Sherrard et al., 2002; Wagemans & Cluitmans, 2006). In our study sample, 19% had a seizure disorder, and 39% of this subgroup experienced one or more falls. Given that adults with intellectual disability have a high prevalence of seizure disorder, it is critical to educate primary caregivers about the importance of monitoring seizures during the episodes to prevent occurrence of falls.
We identified polypharmacy as a risk factor for falls that was also found in previous studies (Hsieh et al., 2001; Sherrard et al., 2002). Polypharmacy is very common in adults with intellectual disability. The use of multiple medications may be masking underlying modifiable risk factors for falls. More research is needed to identify whether fall occurrence is related to the side effects of certain medications or whether there are other factors that may have stronger associations with falls risk (e.g., lower extremity weakness, balance disorders). For caregivers, carefully monitoring side effects of medications and consistently consulting with health care providers are crucial.
Use of walking aids was a common risk factor for the full sample and those who did not have seizure disorders. Having difficulty lifting/carrying greater than 10 lb was another risk factor for the full sample and having difficulty walking three blocks was for those without seizure disorder. This suggests that adults with intellectual disability who have physical limitations are at an increased risk of falls. An emphasis on risk assessments for individual falls that ask about use of walking aids and ability to walk a short distance or lift/carry more than 10 lb may be an important identifier in the provision of care/support services to individuals with intellectual disability at both individual and organizational levels to reduce the risk of falls. Our findings also indicated having a condition of arthritis or osteoporosis, back pain, and urinary incontinence is related to increased risks for falls in adults with intellectual disability without a seizure disorder or fall-related injuries requiring medical care among the fallers. Urinary incontinence also was found as a risk factor for falls in adults with intellectual disability in a previous study (Finlayson et al., 2010). Arthritis or osteoporosis can cause joint pain, stiffness, and dysfunction, which can lead to a great decline in lower extremity strength, resulting in an increased risk for falls. Hence, promoting physical activities for adults with intellectual disability who have arthritis or osteoporosis with special attention to pain management and strength training may be an effective strategy for reducing falls. Research is needed to assess the efficacy and effectiveness of evidence-based fall prevention programs that have been implemented in the general population on adults with intellectual disability who are at a high risk for falls.
One limitation of this study was the use of retrospective data on falls that might result in recall bias and a higher rate of underreporting. In addition, we did not collect data on causes of falls and locations of falls. Therefore, we could not distinguish whether falls were related to environmental or personal factors, which are important to delineate when developing interventions to prevent falls. Given that almost 25% of the falls required medical care, we found that having osteoporosis, back pain, and urinary incontinence were the risk factors for falls requiring medical care. Further investigation of the consequences of falls is an important area for future research.
The literature on fall risk and fall prevention in adults with intellectual disability is scarce. There is an urgent need for more research examining the mediators and moderators associated with falls in adults with intellectual disability, followed by fall prevention interventions that show promise in reducing the risk of falls in this population. Falls in the general older adult population cause catastrophic injury and death in addition to higher health care costs associated with hospitalization and long-term care. In our findings, 20% of young adults and almost one third of middle-aged adults with intellectual disability experienced one or more falls; in addition, the extended lifespan of adults with intellectual disability will likely increase their risk and rate of falls in their later years. It is important, therefore, to target fall prevention programs among younger adults with intellectual disability to reduce their risk and the rate of falls in older adulthood. A multifactoral fall-risk assessment was recommended by the American Geriatric Society, British Geriatrics Society, and American Academy of Orthopaedic Surgeons, which formed panels on the prevention of falls in older individuals in 2001 and 2011 (American Geriatrics Society, British Geriatrics Society, & American Academy of Orthopaedic Surgeons Panel on Falls Prevention, 2001; American Geriatrics Society & British Geriatrics Society Panel on Prevention of Falls in Older Persons Panel on Prevention of Falls in Older Persons, 2011) before developing intervention programs. No such guidelines and recommendations for prevention of falls for adults with intellectual disability exist; thus, there is a need to develop them.
Authors: Kelly Hsieh (e-mail: firstname.lastname@example.org), University of Illinois at Chicago, Disability and Human Development, 1640 West Roosevelt Rd., Rm. 416, Chicago, IL 60608, USA; James Rimmer, University of Alabama at Birmingham; and Tamar Heller, University of Illinois at Chicago.
Editor-in-Charge: Elizabeth A. Perkins
This research was supported by Special Olympics cooperative agreement #U01DD000302 from the U.S. Centers for Disease Control and Prevention and a grant from the National Institute on Disability and Rehabilitation Research, U.S. Department of Education (Grant H133B080009) to the Rehabilitation Research and Training Center on Aging with Developmental Disabilities: Lifespan Health and Function, University of Illinois at Chicago. The content does not necessarily represent the policy of the U.S. Department of Education, and endorsement by the Federal Government should not be assumed.