Problem, Purpose, and Questions: Evidence-Based Practice Proposal
Effectiveness of Physical Therapy Intervention in Decreasing the Risk for Fails in a Community-Dweliing Aging Population
Barbara Susan Rohinson | Joanne M. Gordon | Scott W. Wallentine | Michelle Visio
Purpose: This clinical investigation examined risk factors for falls and the effectiveness of physical therapy intervention to decrease the risk of falls in a community dwelling aging population.
Sample: A convenience sample of 25 community-dwelling elderly adults ages 65 and older participated.
Methods: Al! subjects participated in an extensive examination and evaluation to identify risk for falls and performance on selected measures of balance and mobility. Subjects were given the opportunity to participate in an exercise pro- gram designed to address impairments or functional disabilities. The control subjects received no intervention; however, they were encouraged to continue witb their normal activities, including exercise.
Findings: Significant differences were found between subjects classified as fall- crs and nonfallers in terms of their performance on selected balance and mobil- ity tests. After pbysical therapy intervention, subjects classified as fallers made significant improvements in their performance on the Berg Balance Scale.
Conclusion: Appropriately designed physical therapy intervention in the form of an exercise program can decrease the risk for falls among a community- dwelling aging population identified as having an increased risk of falls.
All authors work at Southwest Missouri State University in Springfield, Missouri.
Barbara Susan Robinson, DPT, PT, is an Assistant Professor in the Departtnent of Physical Therapy.
Joanne M. Gordon, PhD, RN, CS, is an Ass(xiate Professor in the Department of BionieJicitl Sciences.
Scott W. Wallentine, DPT, PT, is an Assistant Professor in the Departtnent of Physical Therapy.
Michelle Visio, PhD, is an As.sistant Professor in the Department of Psychology.
F alls may have devastating consequences for older adults. Each year, approximately one third of persons over tbe age of 65 experience a fall, or unexpected contact of a
body part with the ground or support- ing surface (Blake et al., 1988; Sattin, 1992). An individual who falls may experience a serious injury, including fracture, soft tissue injury, joint dislo- cation, and mobility impairment dur- ing 15-20^ of falls. Forty percent of hospital admissions for individuals over tbe age of 65 are tbe result of fall- related injuries (Sattin, 1990, 1992). Additionally, fall-related injuries and their consequences are associated
with declining function in activities of daily living (ADLs) (Tinetti et al., 1998) and are the leading cause of death from injuries for these individu- als (Sattin, 1992). Injuries from falls represent a significant public health problem witb annual expenditures of greater than $ 10 billion for the care of bip fractures alone {Sattin, 1992).
Research Question The purpose of this study was to iden- tify individuals at risk for falls, deter- mine specific risk factors for falling, and evaluate the effectiveness of phys- ical therapy intervention to decrease tbe risk of falls in a community- dwelling aging population. Tbe hy- potbeses were (1) tbat there are signif- icant differences between fallers and nonfallers in terms of specific risk fac- tors such as [lerfomiance on selected measures of balance and mobility, posture, preexisting health condi- tions, and the use of medication, and (2) that physical therapy intervention can modify risk factors for falling, and as a result of this intervention, sub- jects identified as fallers will demon- strate improvement on selected meas- ures of balance and mobility.
Review of the Literature Risk Factors for Falls and Characteristics of Fallers Falling is a complex problem witb many potential causes, yet it is a health condition that is highly pre- ventable. Risk factors for falls may be either intrinsic (specific to tbe individ- ual) or extrinsic (related to the envi-
Orthopaedic Nursing - {anuary/February 2002 - Volume 21 • Number 1 55
TaRnigthe Mystery Out of Research Quasi-experimental Design
A s orthopaedic nurses we knowthe reality: Imbalance = Fall = Fracture = Surgery = Dependence. And we know from experience that it takes time and effort to identify per- sons at risk for falling. Time we don't always take. Sometimes it is just easi- er to lump all persons over the age of 65 as "potential fallers"and refer them to a physical therapist for instruction.
To separate out the differences between fallers and nonfallers becomes iristantly complicated by the multitude of compounding vari- ables within human study groups. It is much easier to study animal rat models confined within their labora- tory cages exercising on treadmills, eating premeasured caloric meals, and resting at predetermined inter- vals as the lab lights are dimmed. These rat subjects are free from dis- ability, fast foods diets, the stress of stock market declines, and baby sit- ting grandchildren. But rather than being overwhelmed by all these vari- ables, a study design exists that allows for these human variables: the quasi-experimental design.
We can classify designs into a simple tlireefold classification by ask- ing some key questions. First, does the design use random assignment to gmupa?
if random assignment is used, we call the design a randomized experi- ment or a true experiment. If ran- dom assignment is not used, then we have to ask a second question: Does the design use either multiple groups or multiple ways of measurement? If the answer is yes, we would label it a quasi-experimental design. If no, we would call it a nonexperimental design.
So a quasi-experimental design is one that looks a bit like an experi- mental design but lacks the key ingredient — random assignment. Some scientists refer to them as "queasy experiments" because they give the purists a queasy feeling. With respect to internal validity,
they often appear to be inferior to randomized experiments. But there is something compelling about these designs: taken as a group, they are easily more frequently implemented than their randomized cousins.
Probably the most commonly used quasi-experimental design is the nonequivalent groups design. In its simplest form it requires a pretest and a posttest for treated and com- parison groups. It's identical to the Analysis of Covariance design except that the groups are not created through random assignment.
Authors Robinson et al. chose this design to execute their research questions and determine risk factors that orthopaedic nurses can include in their client assessment. These identifiers can help the nurse make appropriate referrals to a physical therapist.
The jury is still out as to whether quasi-experimental designs can ade- quately control selection bias. It Ls safe to conclude that experimental de- signs are superior in this critical respect. But the many problems asso- ciated with experiments render them impractical for many if not most eval- uations. Quasi-ex[X?ri mental designs are often the best practical approach to take for evaluations of health and education programs.
Do we really need more nursing research? Why bother when we have a nursing shortage and are all so fms- trated by health care cutbacks and insurance pressures? Maybe the answer to the question is that research gives us data to support cost- effective treatment plans that provide financial benefits in the long run.
Research results arm nurse man- agers with financial negotiating ammunition when dealing with the hospital administration. Without these results, the nurse manager is left to defend her patient care plans with anecdotal [>ersonai experiences. While these experiences might be tnae, they are disregarded as war sto- ries in business meetings.
The contrived conversation belowillustrates well the different per- ceptions of nurses and outside con- sultants when it comes to assessing the merits of a program.
Nurse manager: Why do we need to evaluate our physical therapy refer- ral program? We have a good han- dle on what's going on with our program and our patients, and we know we are very successful.
Outside Consultant: Because you never know if it was your program or something else that produced the success you are claiming. , •
Nurse manager: Of course we know it's our program. What else would cause all these people to fall less?
Outside consultant: Maybe the more motivated clients are more compU-, ant? •
Nurse manager: That's nonsense. Anyway, we know we have to have our program evaluated. But why do we need to go through the trou- ble of finding a comparison group to do an evaluation?
Outside Consultant: Let's say that 6 months after clients finish physical therapy, TOH. are falliiig less. Would you consider that proof your pro- gram is a success? What proportion of these individuals might be hav- ing fewer falls now if they hadn't gone through the physical therapy instruction?
Nurse manager: I'm not sure, but it wouldn't be as high as 7Wo, I can tell you that.
Outside Consultant: Well, you don't really know that though. For all you know, S<yiiy might be falling less now if they hadn't taken phys- ical therapy instruction.
Nur.se manager: No way. We are pro- viding a valuable service and are realty helping our clients.
Outside Consultant: That may be so.
56 Orthopaedic Nursing - |anuary/Febniary 2002 - Volume 21 " Number 1
but you haven't proven it. And the insurance companies need to know with certainty how successful you have been. They need to know they are getting bang for their buck.
Nurse manager: They are, I assure you.
Outside consultant: Okay, let's say you are doing some good, that indi- viduals who go through physical therapy instruction are indeed falling less than if they hadn't been instructed. How much of an effect are you having? Would half of them have fallen less anyway? One-third? Two-thirds? You can't know that unless you do an evaluation that includes a comparable group of peo- ple who haven't taken physical tlier- apy instmction.
Nurse manager: Even if half of them had fallen less without the instruc- tion, isn't raising that proportion to 70'K) worth it?
Outside Consultant: 1 don't know. What did it cost for that incremen- tal 20%? And how long will the effects of training last?
Nurse manager: Our program is well worth the money...
Nurses live and work with the patients every day. They care about their program and work hard to make it a success. They strongly believe they are doing a good job, and they understandably resent any implication that they are not.
Evaiuators usually have no con- nection with the program, and, more important, no stake in its survival (which sometimes leads to an under- estimation of the threat that evalua- tions can impose on program man- agement and staff). They know that managers are heavily invested in their program and that a manager's assessment of the program — even one aided by reliable monitoring data — will not be accepted by pro-
gram sponsors as a valid test of whether the program is meeting its objectives and is worth what it costs. And they know that many different factors that are unrelated to the design of the program can affect the outcomes of any social program and can easily lead to unwarranted con- clusions about the program.
Many of us have experienced the "business consultants" who sweep through hospitals to evaluate the health care delivery system. We all know the results — layoff by attrition and hiring freezes throughout physi- cal therapy, nursing, laboratory, and radiology personnel. Research data proving cost-effectiveness is the only weapon we have to explain to the consultants how the health careJ team works. Nursing research results! explain the bottom line.
Beth Lucasey, MA, BSN, RN Nurse Clinician, Osteoporosis Clinic University of Kansas Medicai Center
ronment) and include characteristics such as decreased range of motion, inability to produce adequate joint torque, decreased proprioception, impairment of the visual or vestibular systems, or environmental hazards such as the presence of ice, low light, throw rugs, or lack of grab bars (Gill et al., 1999; Sattin et al., 1998).
Prescription medications are also known to increase the risk of falling, including drugs used to treat depres- sion, anxiety, and hypertension (Blake et al., 1988; CampbeU et al., 1989; Cumming 1998; Leipzig et al., 1999; Uu et al., 1995).
Mobility impairment has been reported in the literature as a major risk factor for falling (Graafmans et al., 1996). Age and physical health, in- cluding the presence of chronic med- ical conditions, musculoskeletal im- pairments, declining strength and joint flexibility, decreased sensation in the lower limbs, visual impairments, or the report of frequent stumbles are also factors that influence balance and may increase the risk for falls (Gehlsen & Whaley 1990b; Herndon et al., 1997; Ivers et al., 1998; Lord et al., 1991; Rohbins et al., 1989; Teno et al., 1990; Wegener et al., 1997).
In their analysis, Tmetd et al. (1995)
determined that cognitive impair- ment, low body mass index, the pres- ence of at least two chronic condi- tions, and impaired balance and gait were associated with increased risk of serious injury associated with a fall. Other studies indicate that gait changes in older adults may be pre- dictors of falling (Gehlsen & Whaley, 1990a; Maki, 1997). Most falls occur because of a complex interaction between intrinsic and extrinsic risk factors, and the risk for falls increases as the number of risk factors accumu- late (Lipsitz et al., 1991).
As a result of research on balance and falls, a number of assessment tools that focus on intrinsic factors for falls have been identified as predictive of fall risk in elderly persons, includ- ing the Berg Balance Scale (BBS), Functional Reach Test (FRT), and the Timed "Up and Go" Test (TU>) (Berg et al., 1989; Duncan et al., 1990; Podsiadlo & Richardson, 1991). These assessment tools assist in the identifi- cation of functional skills that can be modified by fall prevention programs.
Effects of Exercise or Balance Training on Falling Prior research supports the use of struc- tured exercise programs to improve
balance and mobility function, thus reducing the risk for falls or the fre- quency of falls (Lord et al., 1995; Province et al., 1995; Wolf et al., 1996). Shumway-Cook et al. (1997) found that a multidimensional exercise pro- gram can improve balance, mobUity, and decrease fall risk in older adults, as well as enhance their functional ability.
Judge et al. (1993, 1994) reported findings that support the hypothesis that an exercise program emphasizing postural control, moderate resistance training, and walking improves sin- gle-stance balance, although they considered these results to be prelimi- nary findings. A group exercise pro- gram was found to be an effective means to improve performance on several measures of fall risk in a group of elderly women over a period of 12 months (Lord et al., 2001, 1995).
Balance control can be taught to the elderly with evidence of improved functioning. Roberts (1989) reported changes in balance among older adults following a 6-week program of aerobic walking that he attributed to improvements in strength, coordina- tion, and flexibility. Topp et al. (1993) examined the relationship between a dynamic resistance strength-training program and static and dynamic
Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1 57
In a review of 11 randomized,
controlled trials, exercise was
found to be effective in lowering the
risk for falls in selected populations.
measures of balance. Their findings suggest that older adults may demon- strate improved measures of dynamic balance as a result of a 12-week inter- vention, but not at a statistically sig- nificant level.
A 9-week program of physical exer- cise performed twice weekly resulted in improved performance on balance assessments in 70- to 75-year-old sub- jects in a Swedish community (Ledin etal., 1990/91).
In a group of healthy 75 to 90 year- old individuals, balance training led to significant improvement in balance (Wolfson et al., 1996). By contrast, strength training using the same exper- imental group led to significant gains in lower extremity isokinetic strength and single limb stance time, but failed to show improvement in balajice out- come measures. In a review of 11 ran- domized, controlled trials, exercise was found to be effective in lowering the risk for fails in selected populations (Gardner et al., 2000).
Skeietai Alignment and Falling Previous studies that examined spinal curve measurements of elderly fallers and nonfallers demonstrated no signif- icant differences between the groups. W(X)dhall-McNeal (1992) reported a nonsignificant relationship between elderly women with a history of falling and measures of forward lean and tht> racic kyphosis. O'Brien et al. (1997) showed a weak relationship between measures of skeietai alignment in the sagittal plane and performance on the BBS, YKV, and TU> in elderly women. They noted that it was not
possible to determine whether the bal- ance impairments occurred as a result of skeletal deformities or if the skeletal deformities occurred in response to balance impairments.
Methods Subjects A convenience sample of 25 male and female subjects 65 years or older par- ticipated in this project. Participants were volunteers recruited from a com- munity senior center and a senior citi- zen apartment complex. Eligibility was determined by questionnaire that examined medical history, use of med- ications, cTjrrent home environment, activities, a self-reported history of falls during the previous 6 months, score on the Foistein Mini Mental State Exam (MMSE) (Foistein et al., 1975), and the ability to maintain standing with or without an assistive device.
Individuals were excluded from the study if they scored lower than 24 on the MMSE, were unable to stand independently for 60 seconds, or had a medical condition that might im- pact their ability to participate in the study {e.g., vesUbular or CNS patholo- gy, or recent orthopaedic, cardiac, or neurologic diagnosis).
Additionally, individuals who resided in an extended<are or assisted- living facility, or who were under 65, were excluded from the study. Three individuals, one control and two test subjects, withdrew after initiating the study because of health reasons unre- lated to participation in the study and are not included in data analysis.
Classification of Subjects as Fallers or Nonfallers Following evaluation and assessment, subjects in this study were identifieil as fallers or nonfallers, based on the following criteria. Identified as fallers were those subjects who had a self- reported history of one or more falls during the past 6 months or subjects who scored < 45/56 on the BBS or sub- jects who had less than 7 inches of extended reach on the FRT.
Due to the extensive time commit- ment required for participation in the exercise program, subjects were allowed to choose to participate as control subjects or in the exercise group and were not randomly assigned to these groups. Seventeen individuals completed the exercise program with 10 subjects classified as fallers and 7 subjects classified as nonfallers. Five individuals agreed to participate as control subjects, but were encouraged to continue with their normal activi- ties, including exercise. None of the control subjects were classified as fall- ers. Characteristics of the remaining 22 subjects {n =10 fallers and n = 12 nonfallers) are presented in Table 1.
This study used a quasi-exĵ wrimen- tal, nonequivalent control group design that has been reported to be an accept- able alternative to an experimental design when randomization is not pos- sible {Cxwk et al., 1979; Portney et al., 1993). Individuals parficipating in the physical tlierapy intervention attended group physical therapy exercise sessions two times a week for 6 weeks. Ad- ditionally, a home exercise program was designed for each subject in the exercise group. Subjects kept a daily log of exercises completed at iiome, as well as a record of falls for 6 montlis follow- ing completion of the program.
Test Procedures Protocol After receiving a thorough explanation regarding the research project, each volunteer signed an infonned consent statement, in accordance with the Human Subjects Review Committee at the sponsodng institution. The MMSE and questionnaire were then adminis- tered to determine eligibility as de- scribed under subjects. Subjects who participated in the physical therapy intervention were required to have the consent and approval of their physi- cian to participate.
58 Orthopaedic Nursing - |anuary/February 2002 - Volume 21 • Number 1
TABLE 1 Characteristics of the Subject Popuiation
Characteristic Fallers Nonfallers* (n=12)
Statistical test
Age distribution: 65-74 years 75-84 years 85-94 years Average age
Gender (proportion of subjects who are male)
Activity level Exercise < once per week Walk outside < once per week
Anthropometries Height (inches) Weight (pounds)
Home environment Live alone Assistance at home Stairs at home
Frequency of imbalance (by self-report) None Monthly Weekly Daily
Falls History of falls during the past 6 months Injured from fall (at any time)
60% 20% 20% 77.4
30%
60% 40%
64.98 180.7
60% 40% 50%
30% 50% 10% 10%
100% 30%
75% 25%
0 70.7
25%
58.3% 75%
64.33 165.42
50% 50% 75%
66.7% 8.3%
16.7% 8.3%
0 41.7%
x\2) = 2.64^ t (20) =-2.30*''
X\^) = O.79
X'(4) = 6.01 X^(4) ^ 5.67
t (20) = -0.47 t(20) =-0.88
X^(1) = 0.22 X^(l) = 0 22 X^(l) = 1.47
X'(4)=7.82
X^(1) = 5.87* X'O) = O.32
Nonfallers include individuals later identified as control subjects. x^ for age categories was not significant, t test for age means was significant. p<.05. **p<.01.
Individuais eligible to parhcipate in the study were asked to return for further evaluation of blood pressure, height, weight, range of motion, neu- rologic function (including lower extremity reflexes, lower extremity sensation, proprioception, and cere- bellar coordination), and vision. Sagittal plane posture of selected sub- jects was also measured. Subjects were further evaluated with three tests to evaluate balance and mobility during functional activities, the BBS, the FRT, and the TU>
Three investigators were involved
in collection of data. An adult nurse practitioner completed measures of blood pressure, height, weight, neuro- logic function, vision, the TU>, and the FRT. Two physical therapists com- pleted measures of cervical and lumbar active range of motion (AROM), BBS, sagittal plane posture, as well as lower extremity AROM and joint torque.
Ail tests and measures were com- pleted in the same order, designed to minimize the effects of fatigue on per- formance on the funrtional measures of balance. Subjects had a minimum of 10 minutes rest before completion of
lower extremity AROM and joint torque measurements. A variety of mus- culoskeletal assessments were made; however, it is beyond the sco[:)e of this article to discuss all of the results relat- ed to these assessments.
All subjects were reevaluated using identical test procedures carried out by the same investigators at the com- pletion of the 6-week physical therapy intervention. Subjects were asked to report any falls that occurred during this time and during a 6-month peri- od following the completion of the pbysical therapy intervention.
Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1 59
Balance and Mobility Assessment Berg Balance Scale The BBS involves 14 items, represent- ing common everyday functional tasks such as arising from a chair, bending, reaching, and transferring from a bed to a chair, and requires approximately 15-20 minutes to complete. A five- point ordinal scale is used to evaluate a subject's performance (0-4), thus yield- ing a maximum score of 56 points (Berg et al., 1989). The quality of per- formance is rated on several functional tasks, while the time taken to complete a task is measured for other tasks.
A score of < 45 on the BBS has been shovm to be predictive of multiple falls in older adults, with subjects scor- ing below 45 more likely to experi- ence multiple falls (Berg et al., 1989, 1992). The BBS is more specific for identifying individuals who are not at risk for falls and less sensitive for iden- tifying those individuals who fall (Bogle Thorbabn & Newton, 1996).
The BBS has been shown to be a valid and reliable measure of balance in older adults (Berg et al., 1989, 1992). Shumway-Cook et al. (1997) developed a model to predict fall risk and concluded that a score on the BBS in combination witb a self-reported history of imbalance had a high level of sensitivity and specificity for com- munity-dwelling older adults.
Functional Reach Test Tbe FRT, developed by Duncan et al. (1990), is described as a dynamic measure of stability during a self-initi- ated movement and is a measurement of postural control and balance that takes approximately 5 minutes to complete. Using a yardstick secured to a wall at right shoulder height, the subject is asked to place his or her shoulder in forward flexion at 90 degrees, make a fist, and lean as far forward as they can without taking a step, or losing their balance.
Functional reach is defined as tbe difference in inches between an indi- vidual's arm length (measured at the base of the third metacarpal) and his or ber maximum forward excursion while maintaining a fixed base of sup- port in standing.
Weiner et al. (1992) determined that individuals witb a functional reach of less than 7 inches demonstrated lim- itations in mobility skills, had the slow- est gait, were unable to perform one- footed stance or tandem walking, and
All subjects were reevaluated using identical test proce- dures carried out hy the same investigators at the completion of the 6-week physical therapy intervention.
were the most restricted in activities of daily living. This test is a reliable and valid measure of postural control and balance in tlie elderly and is highly sen- sitive to change over time as a result of balance training (Duncan et al., 1990, 1992; Weiner et al., 1992, 1993).
Timed "Up and Go" Test The final measure of balance was a modified, timed version of the "Get Up and Go" Test described by Mathias et al. (1986). The Timed "Up & Go" Test (TU>) takes approximately 1-2 minutes to complete and measures the time an individual needs to rise to standing from an armchair, walk a dis- tance of 3 meters, turn, walk back to the chair, and sit down (Podsiadio & Richardson, 1991). No physical assis- tance is given during the test; howev- er, individuals are allowed to use an assistive device if tbey normally use such a device when walking.
Podsiadio & Richardson (1991) found that individuals who are able to complete the TU> in less than 20 seconds have been sliown to be inde- pendently mobile. Those who score greater than 30 seconds tend to need the assistance of others tor many mobility tasks and typically score in tbe middle or lower third of the BBS. Individuals who take between 20 and 29 seconds to complete the test require additional testing to clarify their functional level. The TU> is considered a reliable test and corre- lates well with more extensive meas- ures of balance such as the BBS.
Analysis of Sagittal Plane Posture The Life Mechanics Institute Mid- Sagittal Contour Gauge (Salem, UT) is designed to outline the curvature of the spine in the mid-sagittal plane. It consists of a rectangular, airtight, chamber that measures 97cm x 36cm X 3cm in height, width, and depth, respectively.
Encased within the chamber are 124 hollow alumijium rods, the ends of which are covered witb soft latex tips. Under resting conditions, each of tbe rods extends slightly out of the chamber through a precisely machined exit hole, and are aligned one above the other vertically.
Under 1-2 psi of nitrogen gas intro- duced into the chamber, the rods are deployed against the subject's soft tis- sue overlying the spinous processes of the vertebral column in the mid-sagit- tal plane. A scanner is then moved ver- tically along tbe full length of the chamber face and measures tbe dis- tance of deployment of each of tbe 124 rods, transferring that information into a computer.
The average radius of each of the three spinal airves, tbe amount of cur- vature for eacb curve, the sagittal index, and the degree of forward lean are cal- culated. Measurement procedures were followed as outlined by the manufac- turer of the device. Reliability and valid- ity studies are currently underway.
Physical Therapy Intervention All subjects were given the opportunity to participate in a 6-week fall preven- tion program consisting of group phys- ical therapy exercise sessions. Ad- ditionally, participants in the exercise program received individualized exer- cise programs designed to address spe- cific impairments or functional deficits tbat were identified during their initial evaluation and assessment.
Exercises were performed twice a week in group physical tlierapy exer- cise sessions and daily by the subjects in their home, as reported on a daily log. Attendance was high, witb all subjects attending an average of 10.6 of the 12 exercise sessions scheduled with >50% of the subjects exercising an additional 2-3 times per week in their homes. Only one subject attend- ed fewer than 8 sessions.
60 Orthopaedic Nursing - lanuary/February 2002 - Volume 21 • Number 1
Each group physical therapy exer- cise session lasted for approximately 50 minutes and included ail of the foi- lowing: exercises to improve postural alignment and axial extension, thera- peutic exercise designed to address strength deficits and to improve bal- ance skills, and flexibility exercises for those individuals shown to have impairments in spinal or lower extremity range of motion. The exer- cise sessions also incorporated static and dynamic balance activities.
All exercises could be completed sitting or standing to facilitate the subjects' ability to complete similar exercises in their home. Cervical spine and shoulder girdle flexibility, tho- racic extension, weight shifting, and trunk rotation exercises were complet- ed in a sitting position. These were fol- lowed by ankle flexibility and sitting thigh extension exercises.
Standing hip extension exercises were performed unilaterally with sub- jects stabilizing their positions by holding onto tbe hack of a chair. Unilateral hip abduction exercises fol- lowed. Both of these exercises empha- sized the ability to shift weight from side to side while still maintaining balance. Spinal alignment was em- phasized during these exercises to encourage improved posture with increased lumbar lordosis, decreased thoracic kyphosis, and decreased for- ward lean.
Subjects then moved to an area that allowed them to use a wail for sup- port and completed wall slides to strengthen bilateral quadriceps mus- cles. Subjects then stood approximate- ly 12 inches from the wall (facing the wall) with palms against the wall shoulder width apart and leaned for- ward while rising on their toes and looking up. This exercise allowed stretching of bilateral hip flexors and strengthening of bilateral plantarflex- ors; it also promoted lumbar extension.
Exercises completed in standing position were concluded with stretch- es for plantarflexors. Dynamic gait activities followed, requiring the sub- jects to alter the size of their base of support and increase awareness of the position of their feet during tandem walking, braid walking, and walking forward, backward, and sideways.
During the exercise sessions, sub- jects often requested information regarding activities that they found increasingly difficult with age or were
During the exercise sessions, subjects often requested information regarding activities that they found increasingly difficult with age or were no longer able to complete.
no longer able to complete. As a result, several exercises were incorporated into the program, including the task of rising from a prone or supine position on the floor to a standing position without assistance. All participants, many of whom stated that they had been unable to get up off the floor independently for the last 2 years, were able to leam to rise from the floor.
Another task that was identified as challenging included rising from over- stuffed couches or chairs. Subjects were taught basic transfer techniques to move from sitting to standing with the addition of using the legs and hips to move to the front of the sitting surface.
Other activities that were difficult and often led to loss of balance included reaching into upper level kitchen cabinets. Participants were taught how to use counters as sup- port, rising on the toes of one foot, and extending the opposite arm to reach. Education followed regarding the safe use of step stools, nightlights, and grab-bars, as well as the potential hazards of throw rugs.
Data Analysis Statistical analysis was performed using SPSS for Windows, version 10.0. Descriptive statistics were used to describe all demographic data (see Table 1). For the analyses conducted to examine risk factors for falls, the group identified as nonfallers {/; = 12) includ- ed the control subjects (n = 5) who did not participate in the exercise program as well as subjects identified as non-
fallers (H = 7) who subsequently partic- ipated in the exercise program. These two groups were combined for the risk factor analyses because subject assign- ment to nonfaller exercise group and control group was made after preexer- cise measures were taken. Other than history of falls, statistical tests con- firmed that no significant differences existed in the demographic data between subjects identified as fallers and nonfallers.
Additional analyses were conduct- ed to examine risk factors for falls and the effectiveness of the intervention in improving scores on selected tests of balance and mobility and changing forward lean. A probability value of less than .05 was considered statistically significant unless otherwise indicated.
To assess the effectiveness of the intervention in modifying risk factors for falls, preexercise and postexercise measures were examined using paired-samples t tests and analysis of variance (ANOVA). For these analyses, the nonfaller preexercise group was split into two groups, nonfallers (n = 7) who participated in the exercise group (nonfallers exercise group) and control subjects (n = 5) who did not participate in the intervention.
Results Risk Factors for Falls Prior Medical History and Medication Use Only one of the assessed 29 medical conditions emerged as significant between faUers and nonfallers. All of
Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1 61
TABLE 2 Percent of Medical Conditions Reported at Pretest
by Participants Within Each Group
Medical history
Neurologic Stroke
Orthopaedic Neck pain Low back pain Fracture Arthritis loint replacement Fibromyalgia
Cardiopulmonary system High blood pressure Peripheral vascular disease
Endocrine Diabetes
Other Sinus disorder Surgery Cancer tumor
Croup Fallers (n=10)
30%
30% 70% 40%
100% 10% 0
60% 10%
20%
50% 90% 30%
Nonfallers^ (n=12)
16.7%
25% 58.3% 41.7% 41.7%
0 0
58.3% 8.3%
16.7%
16.7% 66.7% 33.3%
x'
0.55
0.07 0.32 0.01 8.56** 1.26 -
0.01 0.02
0.04
2.79
0.03
Note: df= 1 for all tests. * Nonfallers Include individuals later identified as control subjects.
*p<.05. - p < . 0 1 .
the fallers reported arthritis compared to only 41.7% of the nonfallers (see Table 2).
A significantly greater proportion of fallers {33.3'M)) reported using pre- scription analgesics when compared to nonfallers (O'K.) (see Table 3). Although frequency of use between the two groups of the remaining med- ications did not differ significantly at a probability level of .05 level, two of the categories were significant at a probability level of p < .10. A greater proportion of fallers reported using over-the-counter pain, sleep, and/or sinus medications (66.7*Mi) than non- fallers (25'/f)). Similarly, a greater pro- portion of fallers (22.2'K>) reported using some other over-the-counter medication when compared to the nonfallers (0%).
Neurologic Examination No significant differences emerged between fallers and nonfallers in tests of proprioception, reflexes, sensation, and cerehellar coordination.
Balance and Mobility Assessments Alttiougii Individuals classified as fail- ers required longer to complete the TU>, no statistically significant dif- ferences were found between the fall- ers and nonfaliers on the preexercise measure (see Table 4).
Significant differences were found between the two groups on the preex- ercise BBS and the FRT On the BBS, fallers scored significantly lower than the nonfallers, with the mean of 48.80 approaching the cut off score of 45 to be classified at increased risk for falls.
nonfallers were able to reach signifi- cantly farther on the FRT than fallers. The mean functional reach for fallers, prior to intervention was 6.62 inches, slightly below the 7 inches required to be classified as a nonfaller.
Berg Baiance Scaie There were no statistically significant differences between the nonfallers exercise group or the control subjects on the preexercise and postexercise BBS. However, there was a statistically significant difference in preexercise and postexercise BBS scores for the fallers. As shown in Figure 1 and Table 5, the mean score for the fallers increased from 48.80 to 52.90.
One-way ANOVAs were conducted to examine differences among the groups for postexercise measures and revealed no significant differences among the groups for the postexercise BBS scores, tlius further supporting the positive effect of physical therapy intervention for fallers. TWo of the 17 subjects who participated in the exer- cise program scored less than 45 points on the BBS, scoring 38 and 39 points, placing them at increased risk for falls hefore participation in the exercise program. After 6 weeks of physical therapy intervention, their scores im- proved to 56 and 45, respectively.
Functional Reactt Test No statistically significant differences were found between the preexercise and postexercise FRT scores for fallers or nonfallers exercise group. There was a statistically significant difference in the preexercise and postexercise scores for the control group. The mean score for the controls increa.sed from 10.56 inches to 13.89 inches, as shown in Figure 2 and Table 5. Because of the small number of subjects in this group (n = 5), caution is warranted for inter- preting this increase. A practice effect may explain the increase.
A one-way ANOVA revealed that fallers and nonfallers exercise group differed significantly from controls when measured postexercise, with the control subjects able to reacii signifi- cantly farther than both groups. Six of the 17 subjects who participated in the exercise program demonstrated a functional reach of less than 7 inches. After completion of the exercise pro- gram, only 3 of the subjects had a functional reach of less than 7 inches.
62 Orthopaedic Nursing - |anuary/February 2002 - Volume 21 • Number 1
TABLE 3 Percent of Medications Used at Pretest by Participants Within Each Group
Hypertension
Respiratory Diabetes
Analgesic
Anticoagulant Seizure
Psychotropic Hypnotic Optic
Herbal
Vitamins/minerals Over-the-counter - pain/sleep/sinus Other prescription
Other over-the-counter
Fallers Cn = 9)
66%
22.2% 22.2%
33.3% 11.1%
0 11.1% 11.1% 11.1%
22.2% 66.7% 66.7%
55.6% 22.2%
Croup Nonfallers' {n=12)
66.7%
8.3% 16.7% 0
8.3%
8.3% 8.3% 0
8.3%
8.3% 66.7%
25.0% 33.3% 0
X̂
0.00
0.81 0.10 4.67*
0.05
0.79 0.79 1.40 0.05
0.81 0.00
5.45t 1.04
2.95t
Note: df= 1 for all tests. ^ Nonfallers include individuals later identified as control subjects. tp<.10. *p<.05.
Measure
TABLE 4 Preexercise Baiance and Mobiiity Assessments
Berg Balance Scale Timed "Up and Go" Test Functional Reach Test
*p< .05 . * *p< .01 .
M
48.80 11.50 6.62
Fallers (n=10)
5 4 2
Croup
fD
.85
.30
.23
Nonfallers (n=12)
M
54.83 9.67
10.43
SD
1.59 2.27 2.12
3 -1 4
t
.17**
.28
.11**
Timed "Up and Go" Test No statistically significant differences were found among the scores for fallers, nonfallers exercise group, or the control subjects in the preexercise or postexer- dse mean scores on the TU>. Post- exercise comparisons of the groups revealed fallers and nonfallers exercise group each required significantly longer to complete the test tlian did control subjects (see Figure 3 and Table 5). Only
one subject was unable to perform the TU> in less than 20 seconds before the exercise program; however, she was able to complete the test in 15 seconds during the follow-up assessment.
Analysis of Sagittal Plane Posture No statistically significant differences were found between the faller and nonfallers before the intervention in terms of forward lean as measured in
Clinical Implications Fall-related injury is a major cause of disabilit)' and even death in the elderly. Falling is a complex prob- lem with many potential causes, yet is a health condition that is highly preventable. Identifying those indi- viduals at risk for falls and imple- menting fall prevention programs can help decrease the risk for falls among a community-dwelling aging population.
Individuals at increased risk for falls can be identified by using assessment tools such as the Berg Balance Scale, Functional Reach Test, and Timed "Up and Go" Test. These tests are relatively quick and easy to administer in a variety of settings, require minimal equip- ment, and help to provide a com- prehensive assessment of fall risk.
Research has demonstrated that a variety of exercise interventions, including muscle strengthening, flexibility exercise, and balance training can reduce the risk for falls. Individuals who participated in a 6- week fall prevention program, with exercises performed twice a week in group physical therapy exercise ses- sions and individual exercises per- formed 2 to 3 times a week by sub- jects in their homes, demonstrated a decreased risk for falls.
Nurses and physical therapists can help the elderly maintain their independence and reduce their risks for falls by teaching them cor- rect transfer techniques when mov- ing from a sitting to standing posi- tion, rising from the floor inde- pendently, and modifying activities of daily living.
Collaboration between nurses and physical therapists can be used effectively to identify those at risk for falls and provide intervention to reduce that risk. Nurses, who have more contact with the general pop- ulation than pliysical therapists, could administer balance assess- ments, or recommend that a patient be examined further based on their findings during a patient interview.
Orthopaedic Nursing - lanuary/February 2002 - Volume 21 • Number 1 63
Berg Balance Scale
Preexercise
Postexercise
Controls (n = 5) Fallers ( n = 10)
Subjects
Nonfallers EG (n = 7)
FIGURE 1
A statistically significant difference was found in the preexercise and post- exercise scores of fallers, ((9)=-2.47, p < .05. There were no statistically significant differences for the control or nonfaller exercise groups. There was no statistically significant difference among the groups for postexercise BBS
scores.
Functional Reach
10 -
10-
8
6
4
I
n-
_ —^ 1_• Hffl
•
u•L
Preexercise
Postexercise
Controls (n = 5) Fallers (n = 9)
Subjects
Nonfallers EC (n = 7)
FIGURE 2
The control group showed a statistically significant difference in functional reach preexercise and postexercise, t (4) = -4.30, p < .05. No statistically significant differences were found in those fallers and nonfallers who partic- ipated in the exercise program. Fallers and nonfallers in the exercise group were significantly different from controls when measured postexercise, F(2,18) = 11.62, p < . 0 1 .
ii,n = 7.80, - 5.68,
degrees 3.31; M = 2.15, /
However, individuals catego- rized as fallers tended to have a higher degree of forward iean with a larger standard deviation than nonfallers. In ali other measures of sagittal plane pos- tiire including cervical, thoracic and lumbar depth to length ratio and the ratio of length of the thoracic spine length to lumbar spine length, there were no sig- nificant differences between the preintervention faller and non- faller groups.
There was a statistically sig- nificant decrease {p < .05) for the nonfallers exercise group for- ward lean postexercise. There was no significant difference in fallers postexercise or the con- trol subjects (see Figure 4 and Table 5).
Model for Fall Classification A predictive model for fall risk was constmcted by conducting a forward stepwise logistic regres- sion analysis. The variables con- sidered for the model were deter- mined from the analysis of indi- vidual risk factors that were found to show significant differ- ences between fallers and non- fallers {i.e., history of arthritis, use of prescription pain medica- tion, BBS, FRT, history of falls).
Clearly, the variables used to classify subjects as fallers and nonfallers (BBS, FRT, history of falls) should emerge as impor- tant predictors for fall risk; how- ever, this model was constructed to examine if other subject char- acteristics may also be impor- tant in predicting fall risk. History of falls was not included in model selection because his- tory of falls predicted fall risk perfectly and thus was not of interest as a variable.
The final model, shown in Table 6, included both the both the BBS and the FRT. History of arthritis and use of prescription analgesics did not emerge as sta- tistically significant variables in the model.
64 Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1
Discussion Timed Up and Co Test
It is evident from our results that fallers take significantly more prescription analgesics than non- fallers. The need for prescription analgesics may be due to the increased frequency of a diagno- sis of arthritis or degenerative jointdisease (100̂ )̂ ) of the fallers). Wegener et al. (1997) noted that balance deteriorates in individu- als with knee osteoarthritis.
Typically, individuals with arthritis modify their activity lev- els secondary to complaints of pain associated with this diagno- sis. Decreased activity levels may subsequently lead to decreased lower extremity strength and range of motion. Such changes may have affected the perform- ance of subjects classified as fall- ers on the BBS, FRT, and TU>
Additionally, individuals who were fallers had a tendency to take more medications than nonfallers. This finding is con- sistent with the literature. Gen- erally, fallers have been found to take more medications than nonfallers (Lipsitz et al., 1991). Leipzig et al. (1999) noted that individuals taking more than three to four medications have been shown to have increased risk for recurrent falls.
The nonfallers who were members of the exercise group (n =7) made the most clinically significant change by decreasing their degree of forward lean. These subjects may have been able to achieve a more upright position secondary to the physi- cal therapy intervention.
Alternatively, they may have simply tried harder to "stand up straight" at the postexercise assessment. Failure of the faller group to decrease forward lean may be associated with this group's high report of arthritis (100%) and degenerative joint disease, which could hinder their ability to increase spinal mobility.
The mean forward lean of nonfallers after physical therapy intervention was 5.2 degrees, while fallers had a mean forward lean of 7.7 degrees. The relation- ship between forward lean and risk for falls should be explored
• Preexercise
D Postexercise
Controls (n = 5) Fallers (n = 10) Nonfallers EC (n = 7)
Subjects
FIGURE 3
When preexercise and postexercise scores were analyzed, no statistically significant differences were found for controls, fallers, or the nonfallers in the exercise group for the Timed Up and Go Test. Postexercise, fallers and nonfaliers differed significantly from controls with both groups taking a longer time to complete the test, F (2,19) = 9.07, p < .01.
Forward Lean
Preexercise
Postexercise
Controls (n = 4) Fallers (n = 8)
Subjects Nonfallers EG (n = 5)
FIGURE 4
Preexercise and postexercise measures of forward lean for nonfallers in the exer- cise group were statistically significantly different, t (4) = 3.64, p < .02. There were no statistically significant differences for the controls or fallers on the preex- ercise and postexercise measures of forward lean. There were no statistically significant differences among the three groups postexercise, although the failers and nonfallers in the exercise group had a greater degree of forward lean.
Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1 65
TABLE 5 Balance and Mobility Assessments and Forward Lean
Measure Preexercise Postexercise
Group M SD M SD
Fallers NonfallersEG Controls
Berg Balance Scale
48.80 5.85 54.86 2.04 54.80 0.84
52.90 55.43 55.60
3.11 0.79 0.55
-2.47* -0.66 -2.14
1.15 1.17 0.67
Fallers NonfallersEG Controls
Timed "Up and Go" Test
11.50 4.30 12.70^ 2.63 10.57 2.44 10.86'' 2.34 8.40 1.34 7.40^ 0.89
-0.85 -0.55 1.12
1.60 1.47 0.27
Fallers NonfallersEG Controls
Functional Reach Test
6.66 2.36 10.34 1.84 10.56 2.69
7.30=" 10.17^ 13.89"
2.74 1.76 2.76
-1.08 0.22 -4.30*
1.20 1.25 0.61
Failers NonfallersEG Control
Forward Lean
7.80 3.31 6.10 2.46 5.15 1.89
7.70 ••&M:
3.93
3.74
2.31
0.14 3.64* 0.98
0.03 0.34 0.60
Note: NonfallersEG = Nonfallers Exercise Group. '"'One-way ANOVAs were conducted to examine differences among groups for postexercise measures. Different superscripts indicate
significant differences among groups; Berg Balance Scale f (2,19) = 3.82, p > .05; Timed "Up and Go" Test f (2,19) ^ 9.07, p < .01; Functional Reach Test F (2,18) = 11.62, p < .01; Forward Lean f (2,15) = 2.55, p> .05. Degrees of freedom for Berg Balance Scale, Timed "Up and Go" Test, Functional Reach Test t tests = df fallers = 9. df nonfallersEG = 6. df controls = 4. Degrees of freedom for Forward Lean (test ^ df fallers = 7. d/" nonfallersEG = 4. df controls = 3.
* p < . 0 5 . * * p < . 0 1 .
in more detail with a larger subject population.
During the postintervention assess- ment, all subjects scored 45 or above on the BBS, indicating tbat they are less likely to experience a fall. Ad- ditionally, the standard deviarion for fallers decreased preexercise and pos- texercise signifying a more homoge- neous group. Increased BBS scores for fallers may be due to the effects of exercise on lower extremity flexibility or strength. Studies to evaluate these effects are ongoing.
After the intervention, the mean funcrionai reach for fallers increased from 6.6 inches to 7.3 inches, indicat- ing that as a group they are less likely to demonstrate limitations in mobili- ty or restrictions on activities of daily living. Three individuals remained
classified as fallers based on their FRT. The average age of these individuals was 87.3 years, possibly indicating greater difficulty with this skill with aging. Duncan et al. (1990) noted an influence of age on functional reach with a .7 inch decrease in functional reach with a 10-year increase in age. Additionally, the forward lean of the two oldest members (89 years and 92 years) of the faller group averaged 9.5 degrees, already stressing their limits of stability, and perhaps preventing them from reaching greater than 7 inches during the FRT.
Results of this study demonstrate that physical therapy intervention can reduce the risk for falls for those elderly individuals who are consid- ered to be at increased risk for falls based on scores on the BBS, FRT, and
history of falls. Before the physical therapy intervention, 10 subjects were classified as fallers. After the fall pre- vention program, only four individu- als met the criteria to be classified as fallers, based on their scores on the FRT and their history of falls during the 6 months following completion of the program.
Sixty percent of subjects classified as fallers before the 6 week physical therapy intervention held twice week- ly demonstrated improvements that enabled them to no longer be classi- fied as fallers. It is possible that a longer length of intervention may have decreased tbe risk for falls for a larger percentage of participants. The ideal frequency and duration of an exercise program, as well as the ideal structure of such a program, required
66 Orthopaedic Nursing - lanuary/February 2002 - Volume 21 • Number 1
Risk Factor
Functional Reach Test
Berg Balance Scale
Constant
TABLE 6 Logistic Regression Model
Model Coefficient
-0.79
6.57
62.78
for Faiis
Standard Error
0.34
2.94
33.81
P
.02
.03
.06
for individuals with impairments in balance to demonstrate improvement on balance measures has not been established. However, significant improvements have been noted with exercise programs lasting from 6 to 24 weeks with 1-3 sessions per week {Ledin et al., 1990/91; Shumway- Cook et al., 1997; Taaffe et al., 1999).
After completion of the exercise program, subjects were interviewed by a researcher not involved in the exer- cise program and asked to comment on their perceptions regarding the value of the physical therapy interven- tion. The majority of participants stat- ed the exercise program was of great value to them and has allowed them to remain more active, as well as com- plete tasks that they have not been able to do, in some cases, for over 2 years.
During a 6-month follow-up, 21 of the 22 subjects completing the study were contacted. Two additional sub- jects who had withdrawn early in the study were also contacted. Only one of these subjects had experienced a fall that she attributed to slipping on the ice. Many of the participants in the exercise program provided anec- dotal testimony regarding the impact of the fall prevention program, report- ing increased confidence that they would not fall and the feeling that if they did fall they would be able to independently return to standing.
The camaraderie and supportive interaction among the members of the exercise group appeared to gener- ate interest in participation and con- tinuation of the exercise program. Continuation of the twice-weekly exercise sessions was organized by one participant. These sessions were held in the exercise facility of the senior cit- izen apartment complex. Over 75'Ki of the original participants participated in these group sessions.
Tlie majority of the participants
The majority of partici- pants stated the exercise program was of great value to them and has allowed them to remain more active, as well as complete tasks that they have not been able to do, in some cases, for over 2 years.
stated that they continued to partici- pate in exercise in some capacity at least 2-3 times a week. One subject, identi- fied as a faller during the initial assess- ment who subsequently withdrew from the program secondary to an unrelated health problem, noted a dra- matic change in her frequency of stum- bles that she attributed to learning, dur- ing her short enrollment period, to walk with a wider base of support.
The team efforts between a profes- sional registered nurse and physical therapists allowed the expertise of both groups of practitioners to be used to enhance the functional ability of the participants. A collaborative approach with role interaction between nurses and physical therapists may be useful in identifying individuals who have increased risk for falls.
The BBS, FRT, and TU> are easy and quick to administer, require little equipment, and help provide a com- prehensive assessment of fall risk (Whitney et al., 1998). Functional
tests of balance have the advantages of ease of administration, low cost, and more directly interpretable func- tional relevance. However, they may not offer as much precision or ability to measure subclinical balance impair- ments (Bergetal., 1992).
Nurses, who have more contact with the general population than phys- ical therapists, could administer bal- ance-screening tests in a variety of set- tings, or recommend that a patient be further evaluated based on their find- ings during the patient interview. As a result of these screenings, more individ- uals who are at increased risk for falls could be given the opportunity to par- ticipate in fall prevention programs conducted by physical therapists.
Limitations of the Study A major limitation of the study was the small number of subjects in each group. Because participation in the study required active interest of the subjects, individuals may have self- selected into the fall prevention pro- gram because of a history of falling, frequent stumbling, or the perception that their physical strength or mobili- ty had declined. Therefore, members of the intervention group may consti- tute a select group of the elderly.
Further investigation with larger numbers of subjects and random assignment of subjects to test groups will aid in determining the efficacy of physical therapy intervention in the reduction of an individual's risk for falling.
A second limitation of the study was that portions of the preinterven- tion and postintervention evaluation for each subject were performed by the physical therapist primarily responsi- ble for the group physical therapy ses- sion, thus introducing the possibility of evaluator bias. At the time of the
Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1 67
postintervention evaluation, none of the preintervention scores were avail- able to the investigators, reducing the possibility of evaluator bias.
Implications for Future Research Future research using a larger study population will allow for more gener- alization of the results of this study. Several factors may have influenced the results of this study, including changes in lower extremity AROM and joint torque production, as well as changes in spinal mobility in all planes of movement that may have occurred because of the physical ther- apy intervention.
Further examination of spinal and lower extremity active range of motion measurements and lower extremity joint torque is currently ongoing. An examination of the relationship be- tween sagittal plane posture as meas- ured on a reliable device with a larger subject population may indicate a relationship between the degree of forward lean and risk for falls, as well as accurately describe the spinal cur- vature of an elderly population.
Conclusion The identification of specific risk fac- tors for falls may allow health care professionals to more easily identify individuals at increased risk for falls and provide appropriate physical therapy intervention to prevent falls in a community-dwelling aging popu- lation, ft is possible that an exercise program tbat combines strengthening and dynamic balance activities pro- vides the most appropriate interven- tion to decrease the likelihood of falls in this population.
References Btry, K.O., Maki, B.E., Williams, j.l,,
Holliday, P.J., & Wood-Dauphinee, S.L. (1992). Clinical and laboratory measures of postural balance in an elderly population. Archives of Physical Medicine and Rehabilitation, 7:i, 1073-1 OHO.
ik'rg, K.O., Wo(Hi-Dauphinee, S.L., Williams J.l., & Gayton, D. (1989). Measuring balance in the elderly: Preliminary development of an instrument. Physiotherapy Canada, 41, 04-311.
Blake, A.J., Morgan, K., Bendall, M..|., Dallosso, H., Ebrahini, S.B.J., & Arie,
TH.D., et al. (1988). Falls by elderly people at bome: Prevalence and asso- ciated factors. Age and Ageing, 17(6), 365-372.
Bogle Tbornbabn, L.D., & Newton, R.A. (1996). Use of tbe Berg Balance Test to predict falls in elderly persons. Physical Tlierapy, 76, 576-585.
Campbell, A.J,, Borrie, M.J.. & Spears, G.F. (1989). Risk factors for falls in a com- munity-based prospective study of people 70 years and older. Journal of Gerontology: Medical Sciences, 44{4}, M112-117.
Cook, T.D., & Campbell, D.T. (1979). Quasi-experimentation: Design & analy- sis issues for field settings. Boston: Hougbton Mifflin Company.
Cumming, R.G. (1998). Epidemiology of medication-related falls and fractures in tbe elderly. Drugs Si Aging, 12(\), 43-53.
Downton, J.H., & Andrew, K. (1991). Preva- lence, characteristics and factors asso- ciated witb fails among tbe elderly living at bome. Aging, 3(3). 219-228.
Duncan, P.W., Studenski, S., Chandler, J., 6c Prescott, B, (1992). Functional reacb: predictive value in a sample of elderly male veterans, louniul of Gerontology: Medical Sciences, 47(3), M93-M98.
Duncan, P.W., Weiner, D.K., Cbandler,J., & Studenski, S. (1990). Functional reach: A new clinical measure of bal- ance, journal of Gerontology: Medical Sciences, 45(6), M192-M197.
Folstein, M.R, Folstein, S.F.., & McHugb, P.R. (1975). "Mini-mental state." A practical metbod for grading the cog- nitive state of patients for tbe clini- cian, lournal of Psychiatric Research, 12(3). 189-198.
Gardner, M.M., Robertson, M.C., & Camp- bell, A.|. (2(X)()). Exercise in prevent- ing falls and fall related injuries in older people: A review of randomized controlled trials. British Journal of Sporfi Medicine, 34(1). 7-17.
Geblsen, CM., & VVhaley, M.H. (1990a). Falls in the elderly: Part I, gait. Archives of Physical Medicine and Rehabilitation, 71, 735-738.
Geblsen, G.M., h Wbaley, M.H. (1990b). Falls in tbe elderly: Part II, balance, strength, and flexibility. Archives of Physical Medicine ami Rehabilitation, 71, 739-741.
Gill, T.M., Robison, J.T., WiUiams, C.S., & Tinetti, M.E. (1999). Mismatches between tbe bome environment and pbysical capabilities among commu- nity-living older persons. Journal of the American Geriatrics Society, 47(1), 88-92.
Graafmans, W.C, Ooms, M.E., Hofstec, H.M.A., Bezemer, P.D., Bouter, L.M., & Lips, P. (1996). Falls in the elderly: a prosjiective study of risk factors and risk profiles. American Journal of
Epidemiology, 143, 1129-36. Herndon, J.G., Helmick, C.G., Sattin, R.W.,
Stevens, J,A., DeVito, C, & Wingo, P.A. (1997). Cbronic medical conditions and risk of fall injury events at bome in older adults. Journal of the American Geriatrics Society, 45, 739-743.
Ivers, R.Q., Optom, B., Cummings, R.G., Mitchell, P, & Attebo, K. (1998). Visual impairment and falls in older adults: Tbe blue mountain eye study. Journal of the American Geriatrics Society. 46, 58-63.
Judge, J.O., Eindsey, G., Underwood, M., & Winsemius, D. (1993). Balance im- provements in older women: Effects of exercise training. Physical Therapy, 73, 254-265.
Judge, J.O., Wbipple, R.H., & Wolfson, E.I. (1994). Effects of resistive and balance exercises on isokinetic strength in older persons. Joiiniul of the American Geriatrics Society, 42(9), 937-946.
Eedin, T,, Kronbed, A.C., Moller, C, Moller, M., Odkvist, E.M., & Olsson, B. (1990/91). Effects of balance train- ing in elderly evaluated by clinical tests and dynamic posturography. Journal of Vcstibiilar Resetirch, 1(2), 129-1.38.
Leipzig, R.M., Cumming, R.G., 6t Tinetti, M.E. (1999a). Dnigs and falls in older people: A systematic review and nieta- analysis: L psycbotropic drugs. Journal of the American Geriatrics Society, 47, 30-39.
Leipzig, R.M,, Cumming, R.G,, & Tinetti, M.E. (1999b). Drugs and tails in older people: a systematic review and meta- analysis: H. cardiac and analgesic drugs. Journal of the American Geria- trics Society. 47, 40-50.
LipsiU, L.A., Jonsson, PV., Kelley, M.M., & Koestner, J,S. (1991). Causes and cor- relates of recurrent falls in ambulato- ry frail elderly. Journal of Gerontology, 46(4), Ml 14-122.
Liu, B.A., Topper. A.K,, Reeves, R.A., Gryfe, C, & Maki, B.F.. (1995). Falls among older people: Relationsbip to medica- tion use and ortbostatic hypotension. Journal of the American Geriatrics S<Ki- ety,43(\()), 1141-1145.
I^rd, S.R., Sberrington, C, & Menz, H.B. (2001). Falls in older people: Risk factors and strategies for prevention. Cambridge: Cambridge University Press.
Lord, S.R., Ward, J.A., Williams, P, & Strudwick, M. (1995). The effect of a 12-montb exercise trial on balance, strength, and falls in older women: A randomized controlled trial. Journal of the American Geriatrics Society, 43(\ 1), 1198-1206.
Lord, S.R., Clark, R.D., & Webster, I.W. (1991). Pbysiological factors associat- ed witb falls in an elderly population. Journal of the American Geriatrics Society, 39, 1194-1200.
68 Orthopaedic Nursing - January/February 2002 - Volume 21 • Number 1
Maki, B.E. (1997). Gait changes in older adults: predictors of falls or indicators of fear? loiirnal of the American Geri- atrics Society, 4S{^), 313-320.
Mathias, S., Nayak, U.S., & Isaacs, B. (1986). Balance in elderly patients: The "get-up and ^o" test. Archives of I'hy.mal Medicine Rehabilitation, 67{6), 387-389.
O'Brien, K., Culham, E., St Pickles, B. (1997) Balance and skeletal alignment in a group of elderly female fallers and nonfallers. journal of Gerontology: Biological Sciences, 52(4), 221-226.
Podsiadlo, D. & Richardson, S. (1991). The timed "Up & Go": A test of basic func- tional mobility for frail elderly per- sons, journal of the American Geriatrics Society, 39, 142-148.
Portney K.G., St Watkins, M.P. (1993). Fownliition.s of clinical research: Appli- cations topractice, Norwalk: Appleton & Lange.
Province, M.A., Hadley, I'.C. Hornbrook, M.C., et al. (1995). The effects of exer- cise on falls in elderly patients, foumal of the American Medical A.^sociation, 272, 1341-1347.
Robbins, A..S., Rubenstein, LZ., Josephson, K.R., Schulman, B.L., Osterweil, D., & Fine, G. (1989). Predictors of fails among elderly people: Results of two population based studies. Archives of Internal Medicine, 149, 1628-1633.
Roberts, B. (1989). Effects of walking on balance among elders. Nursing Research, 38, 180-183.
Sattin, R.W., Rodriguez, J. G., DeVito, C.A., SJ: Wingo, P.A. (1998). Home enviion- montal hazards and the risk of fali injury events among community dwelling older persons, fourna! of American Geriatrics Society, 46{6), 669- 676.
Saltin, R.W. (1992). Falls among older per- sons: A public health perspective. Annual Review of Public Health, 13, 489-508.
Sattin, R.W. (1990). The incidence of fail injurj' events among the elderly in a defined population. American joumai of Epidemiology, 131(6), 1028-37.
Shumway-Cook, A., Baldwin, M., Polissar, N.L., St Gmber, W. (1997). Predicting the probability for falls in communi- ty-dwelling older adults. Physical Therapy, 77, 812-819.
Shumway-CtK)k, A., Gruber, W., Baldwin, M., Si I,iao, S. (1997). The effect of multidimensional exercises on bal- ance, mobility, and fail risk in com- munity-dwelling older adults. Physical Therapy, 77, 46-57.
Taaffe, D.R.. Duret, C., Wheeler, S., & Marcus, R. (1999). Once weekly resist- ance exercise improves muscle strength and neuromuscular perform- ance in older adults. Joumai of the American Geriatrics Society, 47{IO), 1208-1214.
Teno, J., Kiel, D.P., Si Mor, V. (1990). Multiple stumbles: A risk factor for falls in com- munity dwelling elderly, a prospective study. Joumai of the American Geriatric Smiety, 38, 1321-1325.
Tinetti, M.E., & Williams, G.S. (1998). The effect of falls and fall injury on func- tioning in community-dwelling older persons. Journal of Gerontology: Medical Sciences, S3A{2), M112-M119.
Tinetti, M.E., Doucette, J., Claus, E. & Marottoli, R. (1995). Risk factors for serious injury during falls by older f>ersons in the community. Joumai of the American Geriatrics Society, 43{l\), 1214-1221.
Topp, R., Mikesky, A., Wigglesworth, J., Holt, W., & Edwards, J.E. (1993). The effect of a 12-week dynamic resistance strength training program on gait velocity and balance of older adults. The Gerontologist, 33{'i), 501-506.
Wegener, L., Kisner, C, St Nichols, D. (1997). Static and dynamic balance responses in persons with hilateral knee osteoarthritis. Joumai of Ortho- pedic & Sports Physical Therapy, 25(1), 13-18.
Weiner, D.K., Bongiomi, D.R., Studenski, S.A., Dimcan, P.W., & Kochersberger, G.G. (1993). Does functional reach improve with rehabilitation? Archives ofl'liysical Medicine and Rehabilitation, 76, 796-800.
Weiner, D.K., Duncan, P.W., Chandler, J., St Studenski, S.A. (1992). Functional reach: A marker of physical frailty. Journal of the American Geriatric Society, 40, 201-207.
Whitney, S.L., Poole, J.L, ik Cass, S.I'. (1998). A review of balance instru- ments for older adults. American Journal of Occupational Tlierapy, 52(8), 666-671.
Wolf, S.L., Bamhart, H.X., Kutner, N.G., McNeely, E., Coogler, G., & Xu, T (1996). Reducing frailty and falls in older persons: An investigation of Tai Chi and computerized balance train- ing. Joumai of the American Geriatrics Society, 44{S), 489-497.
Wolfson, L, Whipple, R., Derby, C, Judge, J., King, M., Amerman, P., et al. (1996). Balance and strength training in older adults: Intervention gains and Tai Chi maintenance. Joumai of the American Geriatrics Society, 44(5), 114M145.
Woodhall-McNeal, A.P. (1992). Changes in posture and balance with age. Aging 4, 219-225.
Statement of Ownership, Management, and Circulation
(Required by 39 U.S.C. 3685)
Date of filing was October 8, 2001. Orthopaedic Nursing (ISSN 0744- 6020 is published bimonthly at East Holly Ave/Box 56, Pitman, Gloucester County, NJ 08071-0056, with headquarters or business offices of the publishers at East Holly Ave/Box 56, Pitman, NJ 08071-0056. The name and address of the pub- lisher is Anthony J. Jannetti, Inc., East Holly Ave/Box 56, Pitman, N| 08071-0056. Editor is Mary Faut Rodts; Managing Editor is Claudia M. Cuddy, both located at East Hoily Ave/Box 56, Pitman, NJ 08071-0056. Owner of the publication is National Association of Orthopaedic Nurses, East Holly Ave/Box 56, Pitman, NJ 08071-0056. There are no bondhold- ers, mortgagees, or security holders.
Total number of copies printed (aver- age for preceding 12 months) - 10,250; paid outside county mail .subscriptions - 9,227; paid in-county subscriptions - none; sales through dealers - none; other classes mailed - none; total paid circulation -9,227; free outside-county circulation - 393; free In-county - none; other classes mailed - none; free ciistribution out- side mail - 265; total free distribu- tion - 658; total distribution - 365; copies not distrthutcd - 356; total - 10,250; percent paid circulation - 92.2%.
Actual number of copies printed (July/August 2001 issue) - 1(),O(X); paid outside county mail subscrip- tions - 8,893; paid in-county sub- scriptions - none; sales tlirough deal- ers - none; other classes mailed - none; total paid cirailation -8,893; free outside-county circulation - 452; free in-county - none; other classes mailed - none; free distribution out- side mail - 2,050; total free distribu- tion - 751; total distribution - 9.664; copies not distributed - 356; total - 10.000; percent paid circulation ~ 92.2%.
I certify that the above statements made by me are correĉ t and complete.
Robert C. Mcllvaine Circulation Manager
Orthopaedic Nursing ~ |anuary/February 2002 - Volume 21 • Number 1 69