Fall prevention, and the subsequent reduction in fall-related injuries, is critically important for preserving health and independence among older adults. Falling, or fear of falling, can predispose an individual to falls, which may result in self-limited activity, disability, or even death.1 Falls are the leading cause of injury-related deaths and are the most common cause of nonfatal injuries and hospital admissions for trauma among adults aged 65 years or older.2–4
Falls typically result from a complex and interdependent mix of medical and physical factors and can be multifactorial in origin.5,6 The National Institute on Aging reported that several medical and physical risk factors may lead to increased fall risk, including lower extremity muscle weakness; poor balance; postural hypotension; medications; and sensory deficits.7 Tinetti et al8 indicated that a positive correlation exists between the number of risk factors and fall risk. As the number of risk factors increases, there is a subsequent increase in fall risk.
Older adults who have cardiac conditions often have multiple intrinsic and extrinsic risk factors related to fall risk. Increased risk factors, as well as physiological changes or cardiac issues that may be superimposed on the typical aging process, may make this population more susceptible to falls than their healthy counterparts.9,10 Cardiovascular issues have accounted for a large percentage of patients who were seen in the emergency department for unexplained or recurrent falls,11 and Berg et al12 indicated that there is likely an underestimation of the role that cardiovascular abnormalities play in fall risk. Jansen et al13 indicated that cardiovascular disorders such as heart failure, low blood pressure, and arrhythmias are strongly correlated with fall risk and must be considered when evaluating fall risk. Kuys et al14 described how older adults with heart failure may be more susceptible to falls as a result of musculoskeletal pain, polypharmacy, orthostatic hypotension, reduced sensation, and shortness of breath with household activities. In a systematic review, Lee et al15 found that patients with heart failure had 1.86 times greater odds of falling, especially when on medications such as benzodiazepines and digoxin, when compared with the general population of older adults. In addition, certain diagnoses such as pulmonary hypertension and heart valve regurgitation,16 as well as use of nonselective beta blockers,17 were indicative of increased fall risk.
Vascular disease of the myocardium and peripheral vasculature is often superimposed on age-related factors, resulting in a reduction of functional mobility and strength.18 Older adults with cardiac dysfunction often demonstrate a maldistribution of blood flow to the lower extremities during exercise, and this may be a contributing factor of exercise intolerance.19 Heart failure has been implicated in the reduction of lower extremity blood flow and the severity of heart failure dictated the amount of absolute blood flow in the lower extremities.20 Gardner and Montgomery21 described that common cardiac rehabilitation (CR) diagnoses such as peripheral artery disease (PAD) and intermittent claudication were significantly correlated with balance and falls (P > .05).21 Individuals with PAD demonstrated a 73% greater prevalence of falling when compared with the non-PAD control group.21 Vogt et al22 found no association between PAD and falls in older women; however, this study relied on self-report of falls from the participants and did not use objective tests and measures.
Reduced lower extremity blood flow has not been directly linked to fall risk; however, McDermott et al23 indicated that reduced blood flow has been linked to shorter walking distance, slower walking velocity, impaired standing balance, and lower levels of physical activity. Moreover, Papa et al24 reported impairments in standing balance, postural control, and muscle fatigue may increase fall risk; therefore, it is reasonable to hypothesize that reduced lower extremity blood flow may lead to increased fall risk.
Reduction in the ability to perform activities of daily living and a decline in functional mobility have been associated with decreased muscle strength and can be a predictor of fall risk.25 For example, the inability to rise from a chair without the use of arms and reduced lower extremity strength have been found to increase fall risk.5,26 Lack of lower extremity strength may affect an individual's ability to perform activities such as transitioning from sitting to standing, walking, or stair climbing, any of which can significantly impact quality of life and, more importantly, may predict future disability.27,28 Cardiac-specific diagnoses, such as PAD, have been associated with walking difficulty and a reduction in lower extremity strength.29 Phase II CR is a comprehensive medically supervised program allowing individuals with a diagnosis of myocardial infarction, stable angina, heart failure, valvular heart disease, or who have had surgical procedures such as coronary artery bypass grafting and cardiac transplantation, to be monitored and to exercise in a safe environment.30 Specific studies regarding the ability of reduced lower extremity strength and lower extremity blood flow to predict fall risk in patients enrolled in a phase II CR program have not been reported. Older adults admitted to a CR program may have multiple comorbidities and risk factors that could lead to increased fall risk during rehabilitation. Therefore, there is a need to determine tests and measures that would enable practitioners to assess the fall risk of patients with cardiac conditions throughout the stages of rehabilitation to enhance services. Insight into the predictive ability of lower extremity strength, using results from the 30-second chair stand test (30CST), and lower extremity blood flow, using results from the ankle brachial index (ABI) to determine fall risk may expedite the use of appropriate interventions to prevent fall-related injuries during rehabilitation and potentially reduce overall fall risk for these patients.
The purpose of this study was to determine whether a quick screening of functional lower extremity strength, as measured by the 30CST, and lower extremity blood flow, as measured by the ABI, were predictors of fall risk, as measured by the functional gait assessment (FGA), among patients enrolled in phase II CR. At one local hospital, over the past 2 years, several patients have fallen while participating in a CR program, and one of those patients required hospitalization due to injuries. Appropriate fall-risk screening measures need to be found that allow health care professionals in this environment to identify patients who may be at increased fall risk. We hypothesized that (1) there would be a statistically significant positive correlation between lower extremity blood flow, using ABI values, and dynamic balance, using FGA scores; (2) there would be a statistically significant positive correlation between lower extremity strength, using 30CST scores, and dynamic balance, using FGA scores; and (3) lower extremity blood flow, using the ABI and lower extremity strength, using the 30CST would statistically significantly predict the FGA score among participants. Information from this study may be used to identify patients who are at increased fall risk and facilitate intervention to reduce fall risk.
This prospective cross-sectional study enrolled patients who were eligible for phase II CR. Participants were selected from a convenience sample of patients receiving phase II CR services at a local hospital. When admitted to the CR program, patients were given a flyer regarding the study by a nurse in CR. Interested patients were then screened by the primary investigator to determine eligibility. Patients who met the following inclusion criteria were consecutively enrolled in the study from April 1, 2017, to July 13, 2017, if they (1) were age 50 years or older, (2) had no history of neurological disorders affecting the central nervous system including stroke, traumatic brain injury, or spinal cord injury, and (3) were able to read and understand English. Patients were excluded from the study if they (1) were taking medication than could affect balance and reported dizziness due to the medication, (2) were unable to complete a 3-step command, (3) were unable to stand independently for 20 minutes, or (4) had an orthopedic injury, surgery, or a fracture within the last 6 months.
Data collection took place at a local hospital's CR department. Demographic and participant characteristic information (age, race, and sex) were collected from a chart review. All participants were screened for the inclusion and exclusion criteria. Each participant was assigned a unique identification number that allowed data to be deidentified. The following outcome data were collected: lower extremity blood flow as measured by the ABI, functional lower extremity strength as measured by the 30CST, and fall risk as measured by the FGA.
The Institutional Review Board at Missouri State University approved the procedures of this study to protect the rights of the participants. The primary investigator met with each patient who expressed interest in participating in the study to determine eligibility. During the interview, the investigator had the individual follow a 3-step command to determine whether the individual had the ability to complete the testing sequence for the research study. The patient was asked to (1) say “Hello,” (2) tap the arm of a chair 3 times, and then (3) say “I'm ready.” If the individual successfully completed the screening procedure, the remainder of the inclusion and exclusion criteria was reviewed to determine whether the patient was eligible to participate in the research study. At that point, all patients who wished to participate in the research study completed the informed consent process.
The same investigator administered all tests to each participant. Identical instructions on how to complete each test were given to each participant. Tests were administered in the following order for all participants: ABI, 30CST, 5-minute rest break, and then FGA. Randomization of the tests was considered to reduce test order bias; however, it was determined that because of the physical requirements of the 30CST and FGA and potential fatigue effect of the 30CST, the order should remain consistent. The 5-minute rest break was given to each participant between the 30CST and FGA to avoid potential interference with scores on the FGA if the patient was fatigued after the 30CST. Participants wore a gait belt during the 30CST and FGA, which enabled the investigator to assist if balance was compromised, reducing fall risk.
Ankle Brachial Index
The ABI is a noninvasive screening tool used to identify PAD or a lack of blood flow in the lower extremities by comparing systolic pressures in the lower leg to systolic pressures in the upper arm.31 In a clinical setting, the ABI is used to determine perfusion status of the lower extremities. A ratio of 1.0 is considered normal, less than 0.9 indicates lower extremity arterial disease, and a ratio of greater than 1.3 is prognostic of elevated perfusion or incompressible vessels.31 The ABI demonstrates good inter-rater reliability (intraclass correlation coefficient [ICC] 0.423),32 has high specificity (83.3%–99.0%) and good accuracy (72.1%–89.2%) when used to diagnose peripheral arterial disease.33 The presence of PAD with intermittent claudication has been associated with a higher incidence of falls, placing individuals with PAD at increased risk for injury.21 The ABI was administered after the participant had rested in a supine position for 10 minutes. The investigator calculated the ABI by taking the systolic blood pressure of the posterior tibialis artery in the ankle on both legs and the brachial artery in both arms using a sphygmomanometer and a Huntleigh M2 Doppler ultrasound (Doppler).34 Initially, the sphygmomanometer was placed around the ankle, and the Doppler signal of the posterior tibialis artery was located. The sphygmomanometer was inflated until the Doppler signal disappeared. The sphygmomanometer was inflated an additional 20 mm Hg and then slowly deflated until the Doppler signal reappeared. The highest value obtained for the ankle and was divided by the highest value in the arm.31
Thirty-Second Chair Stand Test
The 30CST is an assessment of functional lower extremity strength and endurance.35 The number of sit to stand repetitions completed by the participant in a 30-second timeframe is counted. If the individual is over half way to a full standing position at the 30-second mark, that repetition is counted. The score for the 30CST is the number of sit to stand repetitions completed in the 30-second timeframe.35 The 30CST has excellent test–retest reliability (0.84 < R < 0.92) and criterion validity (r = 0.77, 95% confidence interval [CI] [0.64, 0.85]) in community-dwelling older individuals aged 60 to 94 years.36 The use of the 30CST allows for the assessment of individuals with a wide array of ability levels. An individual who is unable to complete 1 repetition receives a score of zero, whereas an individual who is able to complete multiple repetitions would receive that score. For some populations, including community-dwelling older adults, this test is preferred over the 5 and 10 repetition chair stand tests that may be too difficult for them to complete.36 The 30CST was administered using a straight back chair with no arms and a seat height of 17 inches.35 Participants were instructed to cross their arms, placing them across their chest. When the investigator was ready to initiate the test, the investigator said “go.” The participant then assumed a full standing position and then sat back down, repeating the procedure as quickly as possible for 30 seconds.
Functional Gait Assessment
The FGA is used to assess an individual's ability to maintain dynamic balance during 10 different gait tasks.37 The maximum score for the FGA is 30 points, and a score of ≤22 indicates a significant fall risk in community-dwelling older adults.38 The FGA demonstrated excellent inter-rater reliability (ICC = 0.93, P < .001)39 and criterion validity when compared with the Berg Balance Scale in community-dwelling older adults (r = 0.84, P < .001).38 The FGA has been established as a reliable and valid assessment for community-dwelling older individuals37; however, its use for fall-risk assessment in the CR population has not been established.
Because of subjectivity when scoring the FGA, reliability of the primary investigator's ability to score the FGA was assessed before data collection. Videos of 5 individuals performing the FGA were scored weekly for 3 weeks, with an average ICC value of 0.998 (95% CI [0.99, 1.00], P < .001). Standardized instructions for each of the 10 activities of the FGA were given before initiation of each activity. After each activity was completed, a score ranging from 0 to 3 was assigned and the total score was determined. After completion of the third station, the primary investigator reviewed the results of the tests and answered any questions. Total time to complete the study was approximately 45 minutes.
Data analyses were conducted using IBM SPSS for Macintosh, Version 24.40 Data screening was used to identify accuracy, missing data, outliers, linearity, and heteroscedacity. No violations of the data assumptions were found. Descriptive statistics were calculated; nominal data were reported as frequencies and percentages; and continuous data were reported as mean values and SDs (Table 1). Normality of the data was assessed using the Shapiro–Wilk test. Bivariate comparisons were conducted on patient demographic variables to look for differences between those at risk for falls. Continuous variable comparisons were conducted using an independent t test. No significant differences were found. Nominal data were compared using a Fisher exact test, and correlations were performed using a Pearson product–moment coefficient of correlation. If a correlation was found to be statistically significant, correlation strength was interpreted using Munro's descriptive terms: r = 0.00 to 0.25: little, if any correlation; r = 0.26 to 0.49: low correlation; r = 0.50 to 0.69: moderate correlation; r = 0.70 to 0.89: high correlation; and r = 0.90 to 1.00: very high correlation.41 A multiple linear regression was used to test the research hypothesis and assess the predictive relationship between the variables to determine whether functional lower extremity strength and lower extremity blood flow could predict the score on the FGA, indicating the presence or absence of fall risk. Variables were introduced into the regression model simultaneously using the enter method. All tests were 2-tailed, and an alpha level of 0.05 was considered statistically significant.
A total of 60 participants were enrolled in the study. A power analysis indicated a minimum of 55 participants were needed to detect a statistically significant effect, if one existed. When verifying the data input, an error was found in an age calculation, and 1 participant did not meet the minimum age requirement. Two participants were unable to complete the testing procedures. Scores from these 3 participants were excluded from the data analyses (n = 57). The final sample of 57 patients receiving CR was predominantly male (70.1%) and white (92.9%), with a mean age of 68.58 years. Of the 57 participants, 61.4% were at risk for falls based on FGA scores.
An exploratory data analysis using a Shapiro–Wilk test revealed that neither the ABI nor the FGA variables significantly deviated from normality beyond P < .05; however, the 30CST was not normally distributed. Residuals met the assumption of independence with a Durbin–Watson value of 2.15 on a scale from 0 to 4. A value near 2 indicates no autocorrelation was found in the residuals from the regression analysis.42 Linearity and homoscedasticity were assessed visually using a plot of standardized residuals against the predicted values. Collinearity statistics indicated that multicollinearity was not an issue (ABI, tolerance = 0.99; 30CST, tolerance = 0.99) and met the assumption of collinearity with a tolerance value of <0.10, with 0.10 being recommended as the minimum level of tolerance, or the nonassociation between 2 variables.43
Results for descriptive statistics for age, ABI, 30CST, and the FGA and correlational analyses are presented in Table 1.
Relationship Among the Ankle Brachial Index, 30-Second Chair Stand Test, and Functional Gait Assessment
Hypothesis 1 stated that lower extremity blood flow, using the ABI, would relate significantly with dynamic balance, using the FGA among participants. A Pearson product–moment correlation coefficient was used to test Hypothesis 1. As shown in Table 1, lower extremity blood flow (ABI) was very weakly and positively correlated with dynamic balance (FGA), r = 0.02, P = .438, and was not statistically significant. Therefore, hypothesis 1 was not supported. The r 2 value of 0.0004 means that only 0.04% of the variance in dynamic balance (FGA) was explained by lower extremity blood flow (ABI).
Hypothesis 2 stated that functional lower extremity strength (30CST) would relate significantly positively with dynamic balance (FGA). Hypothesis 2 was tested using a Spearman's rho correlation because of the non-normal distribution of the 30CST. Lower extremity strength (30CST) was statistically significant and moderately positively correlated, with dynamic balance (FGA), r = 0.71, P < .001. Hypothesis 2 was supported. The r 2 value of 0.46 means that 46% of the variance in the dynamic balance was explained by functional lower extremity strength and indicated a small to medium effect size.44
Procedure and Results for Multiple Regression Test
A multiple regression analysis was used to test hypothesis 3 relating to the relationship between the independent variables of lower extremity blood flow (ABI) and functional lower extremity strength (30CST), and the dependent variable of dynamic balance (FGA). The 2 predictors (ABI and 30CST) were regressed on the dependent variable (FGA) in a forced entry regression analysis. Hypothesis 3 stated that lower extremity blood flow (ABI) and functional lower extremity strength (30CST) would statistically significantly predict dynamic balance (FGA) score among participants. In the final model, the 30CST significantly predicted the overall FGA score and accounted for just over 45% of variance, F(2,54) = 23.97, P < .001, R 2 = 0.47. Analyses of the coefficients revealed that the ABI was largely unrelated to FGA scores (β = −0.05, P = .608); however, the 30CST was a significant predictor of the score on the FGA (β = 0.69, P < .001). Overall, hypothesis 3 was supported. Multiple regression analysis results can be found in Table 2.
Procedure and Results for Binomial Logistic Regression
A binomial logistic regression was performed to ascertain the ability of the 30CST to predict the likelihood that participants were at risk for falls and to generate a receiver–operator characteristic (ROC) curve between chair stand repetitions as the independent variable and fall-risk FGA score of less than or equal to 22 as the dependent variable. The threshold of a 22 or less on the FGA was based on a criterion that has been cited in the literature.38 The logistic regression model was statistically significant, χ2(1, N = 57) = 30.21, P < .001. The model explained 54.9% (Nagelkerke R 2) of the variance in FGA scores and correctly classified 77.2% of the cases. Positive predictive value (sensitivity) was 70.4%, and negative predictive value (specificity) was 83.3%. Higher functional lower extremity strength (30CST) scores were associated with better dynamic balance (FGA) scores, indicating reduced fall risk.
The ROC curve was calculated by plotting the CST-predicted probability of fall risk with a cutoff score of 22 against fall risk measured by the FGA (area under the curve [AUC] = 0.88, SE = 0.045, 95% CI: 0.80, 0.97, P < .001); (Fig. 1). The AUC can be interpreted as the probability that a randomly chosen participant with high fall-risk subject is rated as more likely to be at high fall risk than a randomly chosen participant with low fall risk.45 The predicted probability values were noted to have at least 1 tie between the positive actual state group and the negative actual state group. The cutoff score for CST was 10.5 repetitions of chair stands. This produced the diagonal segments noted in Figure 1.
The purpose of this study was to investigate whether lower extremity blood flow (ABI) and functional lower extremity strength (30CST) could predict fall risk, as measured by the score on the FGA. As expected, functional lower extremity strength (30CST) did predict a significant portion of the score on the FGA (dynamic balance); however, lower extremity blood flow (ABI) was a poor predictor. The results of this study both support and expand previous research; however, literature specific to individuals enrolled in phase II CR is very limited.
In this study, no correlation was found between the ABI and fall risk as measured by the FGA (r = 0.02, P = .438). Moreover, the standardized beta value for the ABI (β = −.05, P = −.516) indicated that the ABI did not significantly contribute to the model and could be removed. Although the ABI was not significantly correlated with scores on the FGA, a definite trend was observed. When examining individual participant scores, as the ratio approached 1.20 or was less than 0.90, the scores on the FGA tended to be lower and most were indicative of fall risk. Ankle brachial index results of less than 0.90 or approaching 1.20 may warrant further testing to rule out fall risk in this population. Results found in the literature regarding lower extremity blood flow (ABI) and fall risk are inconsistent. Gardner and Montgomery21 described that the ABI was significantly correlated with balance and falls in individuals with PAD and intermittent claudication (P > .05).21 Individuals with PAD demonstrated a 73% greater prevalence of falling when compared with the non-PAD control group.21 Vogt et al22 found no association between PAD and falls in older women; however, this study relied on self-report of falls from the participants and did not use objective tests and measures.
Functional lower extremity strength (30CST) was found to be significantly correlated with dynamic balance (FGA). More importantly, the 30CST was found to be a positive predictor of scores on the FGA (β = 0.69, P < .001), accounting for 45.1% of the variability in the FGA (R 2 = 0.47, P < .001), and was a significant contributor to the explanatory power of the model. For every 1 unit of change in functional lower extremity strength, there was a 0.69 unit change in the FGA (dynamic balance) score. Ward et al26 reported a similar finding between functional lower extremity strength using the 5-repetition chair stand test (5CST) and fall risk. A time of greater than 16.7 seconds on the 5CST was an independent predictor of injurious falls in the older adult population. Moreover, the 5CST performed better than the participants' combined score, which included the components of balance and gait speed.26 Tiedemann et al46 not only found that functional lower extremity strength using the 5CST was linked to increased fall risk, but participants who experienced multiple falls had lower scores on the 5CST.
A logistic regression was conducted to determine how the study results could be applied clinically. The 30CST was highly sensitive and specific, indicating that a cutoff score of ≤10.5 repetitions on the 30CST could identify individuals enrolled in CR who were at risk for falls 70.4% of the time and identify those individuals who were not at risk for falls 83.3% of the time. The 30CST demonstrated a moderately high positive predictive value, which would allow clinicians working in CR to predict fall risk using the 30CST as a quick screen with 73.9% accuracy. Reider and Gaul47 reported a similar finding using the 5CST in community-dwelling older adults. In their study, the 5CST test was an effective screening tool that could identify older adults who were at high risk for falls.47
Limitations of this study include that participants in this study were a nonrandom sample of convenience from a single outpatient CR program. This may limit the ability to generalize findings to other patient populations. Studies have shown that comorbidities can increase fall risk in the older adult population.6 Although a quick screen was conducted to determine whether the patient met the inclusion and exclusion criteria for the study (i.e., had chronic medical conditions that may impact test results), data on specific comorbidities were not collected. The outcome assessments used did not have established validity in this patient population. Finally, the mean ABI score was 0.98 (essentially normal) with only 12% having a score of less than 0.90 or greater than 1.20. The lack of ABI scores outside these ranges may have affected the ABI study results.
Results of this study indicate that the ABI was not an effective predictor of fall risk; however, if the ABI scores are less than 0.90 or greater than 1.20, further testing may be warranted to rule out fall risk. In this study, the 30CST predicted a significant portion of the FGA score and could be considered for use as a fall-risk screening measure in the CR population. The 30CST is a very quick, effective, and affordable option and does not require the specialized training needed to administer the FGA nor the ABI. In addition, the 30CST takes less than 2 minutes to perform, making it a very quick and affordable option when compared with the FGA and ABI, which take approximately 20 minutes for each test. By using the cutoff score for the 30CST, once fall risk is determined, professionals working in a CR setting could refer the patient to another health care professional, such as a physical therapist, who could perform a more comprehensive balance evaluation and recommend treatment to reduce fall risk. Future studies should be conducted to determine whether established scores on the 30CST for community-dwelling older individuals can predict fall risk in this population.
1. Toebes MJ, Hoozemans MJ, Furrer R, Dekker J, van Dieën JH. Associations between measures of gait stability, leg strength and fear of falling. Gait Posture. 2015;41(1):76–80.
2. Burns ER, Stevens JA, Lee R. The direct costs of fatal and non-fatal falls among older adults
—United States. J Safety Res. 2016;58:99–103.
3. Carroll NV, Slattum PW, Cox FM. The cost of falls among the community-dwelling elderly. J Manag Care Pharm. 2005;11(4):307–316.
4. Centers for Disease Control and Prevention. Injury prevention & control: Data & statistics (WISQARS). Centers for Disease Control and Prevention: http://www.cdc.gov/injury/wisqars/
. Accessed March 7, 2017.
5. Berry SD, Miller RR. Falls: Epidemiology, pathophysiology, and relationship to fracture. Curr Osteoporos Rep. 2008;6(4):149–154.
6. Dionyssiotis Y. Analyzing the problem of falls among older people. Int J Gen Med. 2012;2012(5):805–813.
8. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med. 1988;319(26):1701–1707.
9. Afilalo J, Karunananthan S, Eisenberg MJ, Alexander KP, Bergman H. Role of frailty in patients with cardiovascular disease. Am J Cardiol. 2009;103(11):1616–1621.
10. Goel K, Shen J, Wolter AD, et al. Prevalence of musculoskeletal and balance disorders in patients enrolled in phase II cardiac rehabilitation
. J Cardiopulm Rehabil Prev. 2010;30(4):235–239.
11. Davies AJ, Kenny RA. Falls presenting to the accident and emergency department: Types of presentation and risk factor profile. Age Ageing. 1996;25(5):362–366.
12. Berg WP, Alessio HM, Mills EM, Tong C. Correlates of recurrent falling in independent community-dwelling older adults
. J Mot Behav. 1997;29(1):5–16.
13. Jansen S, Bhangu J, de Rooij S, Daams J, Kenny RA, van der Velde N. The association of cardiovascular disorders and falls: A systematic review. J Am Med Dir Assoc. 2016;17(3):193–199.
14. Kuys S, van der Ham E, Phty B, et al. Falls and musculoskeletal pain in older adults
with chronic heart failure. Cardiopulm Phys Ther J. 2013;24(4):12–17.
15. Lee K, Pressler SJ, Titler M. Falls in patients with heart failure: A systemic review. J Cardiovasc Nurs. 2015;31(6):555–561.
16. van der Velde N, Stricker BH, Roelandt JR, Ten Cate FJ, van der Cammen TJ. Can echocardiographic findings predict falls in older persons. PLoS One. 2007;2(7):e654–e661.
17. Ham AC, van Dijk SC, Swart KA, et al. Beta-blocker use and fall risk
in older individuals: Original results from two studies with meta-analysis. Br J Clin Pharmacol. 2017;83(10):2292–2302.
18. Lakatta E. Cardiovascular regulatory mechanisms in advanced age. Physiol Rev. 1993;73(2):413–467.
19. Wada O, Asanoi H, Miyagi K, et al. Quantitative evaluation of blood flow distribution to exercising and resting skeletal muscles in patients with cardiac dysfunction using whole-body Thallium-201 scintigraphy. Clin Cardiol. 1997;20(9):785–790.
20. Lindsay D, Holdright DR, Clarke D, Anand IS, Poole-Wilson PA, CollinsP. Endothelial control of lower limb blood flow in chronic heart failure. Heart. 1996;75(5):469–476.
21. Gardner A, Montgomery P. Impaired balance and higher prevalence of falls in subjects with intermittent claudication. J Gerontol A Biol Sci Med Sci. 2001;56A(7):M454–458.
22. Vogt MT, Cauley JA, Kuller LH, Nevitt MC. Functional status and mobility among elderly women with lower extremity arterial disease: The study of osteoporotic fractures. J Am Geriatr Soc. 1994;42(9):923–929.
23. McDermott MM, Greenland P, Liu K, et al. The ankle brachial index
is associated with leg function and physical activity: The walking and leg circulation study. Ann Intern Med. 2002;136(12):873–883.
24. Papa EV, Garg H, Dibble LE. Acute effects of muscle fatigue on anticipatory and reactive postural control in older individuals: A systematic review of the evidence. J Geriatr Phys Ther. 2015;38(1):40–48.
25. Pijnappels M, van der Burg JC, Reeves ND, van Dieën JH. Identification of elderly fallers by muscle strength measures. Eur J Appl Physiol. 2008;102(5):585–592.
26. Ward RE, Leveille SG, Beauchamp MK, et al. Functional performance as a predictor of injurious falls in older adults
. J Am Geriatr Soc. 2015;63(2):315–320.
27. Seo Y, Yates BC, Norman JF, Pozehl B, Kupzyk K. Comparisons of dyspnea, fatigue, and exercise intolerance between individuals with heart failure with high versus low knee extensor muscle strength. Cardiopulm Phys Ther J. 2014;25(1):11–17.
28. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 1995;332(9):556–561.
29. McDermott MM, Fried L, Simonsick E, Ling S, Guralnik JM. Asymptomatic peripheral arterial disease is independently associated with impaired lower extremity functioning. Circulation. 2000;104(4):1007–1012.
30. Price KJ, Gordon BA, Bird SR, Benson AC. A review of guidelines for cardiac rehabilitation exercise programmes: Is there an international consensus? Eur J Prev Cardiol. 2016;23(16):1715–1773.
31. Wound Ostomy and Continence Nurses Society. Ankle brachial index
: Quick reference guide for clinicians. J Wound Ostomy Continence Nurs. 2012;39(2S):S21–S29.
32. Holland-Letz T, Endres HG, Biedermann S, et al. Reproducibility and reliability of the ankle-brachial index as assessed by vascular experts, family physicians and nurses. Vasc Med. 2007;12(2):105–112.
33. Xu D, Li J, Zou L, et al. Sensitivity and specificity of the ankle-brachial index to diagnose peripheral artery disease: A structured review. Vasc Med. 2010;15(5):361–369.
36. Jones CJ, Rikli RE, Beam WC. A 30-s Chair-Stand Test as a measure of lower body strength in community-residing older adults
. Res Q Exerc Sport. 1999;70(2):113–119.
37. Wrisley D, Marchetti G, Kuharsky D, Whitney S. Reliability, internal consistency, and validity of data obtained with the Functional Gait Assessment
. Phys Ther. 2004;84(10):906–918.
38. Wrisley D, Kumar N. Functional Gait Assessment
: Concurrent, discriminative, and predictive validity in community-dwelling older adults
. Phys Ther. 2010;90(5):761–773.
39. Walker M, Austin A, Banke G, et al. Reference group data for the functional gait assessment
. Phys Ther. 2007;87(11):1468–1477.
40. IBM Corp. IBM SPSS Statistics for Macintosh. Armonk, NY: IBM Corp; 2011.
41. Kellar SP, Kelvin EA. Munro's Statistical Methods for Health Care Research. 6th ed. New York, NY: Wolters Kluwer; 2013.
42. Durbin J, Watson GS. Testing for serial correlation in least squares regression, I. Biomerika. 1950;37(3-4):409–428.
43. Tabachnick BG, Fidell LS. Using Multivariate Statistics. 4th ed. Boston, MA: Allyn and Bacon; 2010.
44. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Earlbaum Associates; 1988.
45. Hanley JA, McNeil BJ. The meaning and use of the area under receiver operating characteristic (ROC) curve. Radiology. 1982;143:29–36.
46. Tiedemann A, Shimada H, Sherrington C, Murray S, Lord S. The comparative ability of eight functional mobility tests for predicting falls in community-dwelling older people. Age Ageing. 2008;37(4):430–435.
47. Reider N, Gaul C. Fall risk
screening in the elderly: A comparison of the minimal chair height standing ability test and 5-repetition sit-to-stand test. Arch Gerontol Geriatr. 2016;65(2016):133–13.