Journal of Geriatric Physical Therapy:
Functional Predictors of Stair-Climbing Speed in Older Adults
Hinman, Martha R. PT, EdD1; O'Connell, Janelle K. PT, DPT, PhD, ATC, LAT1; Dorr, Melissa PT, DPT2; Hardin, Robyn PT, DPT3; Tumlinson, Allison B. PT, DPT4; Varner, Bria PT, DPT5
1Department of Physical Therapy at Hardin-Simmons University, Abilene, Texas.
2HealthSouth Inpatient Rehabilitation Hospital, The Woodlands, Texas.
3Dell Children's Medical Center, Austin, Texas.
4Select Physical Therapy, San Antonio, Texas.
5South Texas Rehabilitation Hospital, Brownsville, Texas.
Address correspondence to: Martha R. Hinman, PT, EdD, Department of Physical Therapy, Hardin-Simmons University, 2200 Hickory, HSU Box 16065, Abilene, TX 79698 (firstname.lastname@example.org).
This study was presented as a poster at the American Physical Therapy Association Combined Sections Meeting (Section for Geriatrics) in Chicago, Illinois, on February 2012.
The authors declare no conflicts of interest.
Background and Purpose: Falls on stairs are a common cause of injury and death among older adults. Although stair climbing is a component of some instruments that assess activities of daily living, normal speeds for safe stairway ambulation have not been established. Furthermore, little is known about which components of functional mobility are most highly associated with stair-climbing speed. The purposes of this study were to determine the range of normal stair-climbing speeds for ambulatory, community-dwelling older adults and identify which functional mobility tests could best explain this speed.
Methods: Twenty men and 34 women older than 65 years completed 6 functional mobility tests, including timed heel rises, timed chair stands, functional reach, one-legged stance time (OLST), a timed step test (alternately touching a step 10 times), and self-selected gait speed. Participants were then timed as they ascended and descended a flight of 8 to 10 steps. Combined ascent-descent times were used to calculate stair-climbing speed in steps per second. Stepwise regression techniques determined the best functional predictors for stair-climbing speed.
Results: Participants ascended and descended stairs at an average speed of 1.3 steps per second; men tended to ambulate stairs more quickly than women. The best predictors of stair-climbing speed were usual gait speed and OLST (R = 0.79; P = .01), which explained 63% of the variance in stair-climbing speed.
Discussion: Our results were similar to others who reported stair-climbing speeds ranging from 1.1 to 1.7 steps per second for older adults. However, the 2 predictors identified in this study provide a simpler and more accurate model for estimating stair-climbing speed than has been previously reported. Further research is needed to determine whether this speed is sufficient for negotiating stairs in an emergency. In addition, further study is needed to determine which tests/measures best differentiate individuals who can and cannot independently climb a typical flight of stairs.
Conclusions: An older adult's stair-climbing speed can be accurately estimated by using a model that includes his or her usual gait speed and OLST. This information will help health care professionals and directors of residential facilities make appropriate decisions related to living accommodations for their older adult clients.
Stair climbing is an essential activity of daily living that contributes to one's functional independence and quality of life. Numerous public facilities (eg, theatres, churches, museums, restaurants, shops) still have exterior and/or interior steps that one must climb to access the building. Many older buildings are not equipped with public elevators because of their age and associated cost of renovation. These buildings are not atypical in terms of the types of establishments that older adults might wish to access. Thus, an individual who is unable to ascend or descend stairs safely may potentially put himself or herself at risk when visiting these facilities and negotiating the stairs. In addition, many older adults live in multistory residential buildings where they must be able to quickly negotiate stairs in the event of an emergency, such as a fire, when the elevators may be inoperable. In 1999, the United States Fire Administration1 reported that people older than 75 years had a fire death rate that was 3 to 4 times the national average. They noted that 2 of 3 fire deaths and 1 of 3 fire-related injuries in older adults occurred when they were trying to escape. Mobility limitations contribute significantly to these safety risks for older residents. In their study of patients released from inpatient rehabilitation, Gordon et al2 reported that only 31% were able to negotiate stairs independently.
Approximately 1 of 3 older adults in the United States experience a fall each year.3,4 Although Lord et al5 reported that a relatively small percentage of falls in older adults occur on stairs (∼6%-12%, depending on the age group), these falls often resulted in more serious injuries than those which occurred on a level surface. Startzell et al6 concluded in their review that stair descent is particularly hazardous and is implicated in 3 times more accidents than stair ascent. This conclusion is also supported by a 1988 study by Tinetti et al7 who reported that 75% of the falls reported on stairs occurred during descent. In a survey study by Austrian researchers, a 37% hospitalization rate was associated with falls on stairs; injury risk was notably higher among older adults, particularly women who lived alone.8 A Scottish study investigated 51 cases of death following a fall down stairs; of those cases, 59% went unnoticed until death had occurred.9 Fifty-three percent of these individuals died at the bottom of the stairs, and the rest were hospitalized and died later from complications. The most recent US data published by the National Safety Council10 indicate that nearly 2000 Americans die each year from falls on stairs or steps; this accounts for approximately 8% to 9% of all fall-related deaths in recent years.
Despite the higher risk of fall injury or death associated with stair ambulation among older adults, few activities of daily living instruments specifically assess this functional activity. In a systematic review of 92 functional general mobility scales, van Iersal et al11 reported that only 43 scales included items related to stair-climbing ability, and none of these indicated whether the person could safely negotiate stairs. Although self-reports of stair-climbing ability are most common, these questionnaires are often inaccurate because many older adults either do not want to disclose their limitations or do not perceive that they have stair limitations.12
Several investigators have attempted to correlate various physical characteristics and health conditions with stair-climbing ability.6 Bohannon et al13 noted a higher body mass index among those who could not climb stairs, while others have reported a perception of “poor health.”14 In a survey study, Verghese et al15 asked a group of older adults about their health history and stair-climbing ability, then measured these individuals' step length, cadence, gait velocity, and grip strength. They concluded that individuals who had the greatest difficulty with stair ambulation were those who had less education, variable step lengths, hypertension, osteoarthritis, shorter step length, and weaker grip strength.
Bean et al16 demonstrated that impaired lower limb power was a significant predictor of limited physical performance and functional mobility in community-dwelling older adults. Similarly, Suzuki et al17 reported that plantarflexor power was the strongest predictor of gait speed and dorsiflexor power was the best independent predictor of stair-climbing speed. However, Carmeli et al18 concluded that muscle strength and functional ability were not highly correlated in their study, nor were increases in muscle strength correlated with improvements in functional abilities. In addition to measures of lower extremity strength, Menz and Lord19 examined the influence of foot pain and deformities, sensory loss, reaction time, visual deficits, and balance on tests of functional mobility, which included gait speed, alternate stepping test, stair ascent and descent. They found that participants who had a greater number of foot problems were more likely to experience difficulty with their functional mobility and were more likely to fall.
Despite evidence of various physical and demographic factors associated with stair-climbing ability in older adults, health care professionals still lack a simple method for determining whether older adults can negotiate stairs in a safe and efficient manner. Such a tool would help health care professionals recommend appropriate living environments when discharging patients, as well as identify those who may need additional rehabilitation.13,16,18,20 This information may also be useful to directors of multistory retirement and assisted living facilities who need to consider residents' ability to safely evacuate the building in an emergency. Thus, the purposes of this study were to (1) determine reference values for stair-climbing speed among a group of community-dwelling older adults and (2) develop a valid and accurate prediction model for stair-climbing speed in that population.
Fifty-four community-dwelling older adults were recruited from 4 retirement/assisted living facilities and 2 churches in Abilene, Texas. The sample included 20 men and 34 women with no known pathologies; their ages ranged from 65 to 93 years (mean = 80 (8) years). Participants had to be able to ambulate independently, either with or without an assistive device; those who had more than mild cognitive deficits (as evidenced by an inability to follow basic instructions) were excluded. All participants negotiated stairs routinely either in their homes or local community. This study was approved by the university research review committee at Hardin-Simmons University in Abilene, Texas.
Data Collection Procedures
Before testing, participants signed an informed consent form, provided their age and body weight, and specified their leg dominance (based on which leg they would use to kick a ball). Participants then performed a sequence of 7 functional tests in random order. Six tests of functional mobility were selected to assess lower extremity strength, static and dynamic balance, and gait speed, in addition to the stair-climbing test. Balance and gait tests were performed twice and the better score was used in the data analysis. Participants were allowed to rest as needed between tests. Manual guarding was provided, as needed, by 1 or more investigators during each test.
Tests of functional strength included a heel rise test (number of times the participant could lift the heel off the ground in 30 seconds) and timed chair stands (time needed to stand up and sit down 5 times from a standard chair). Both of these tests have demonstrated good validity and reliability in previous studies.21–28 However, the heel rise test was modified to allow participants to perform it bilaterally because many could not perform the test independently using just one leg as described in the protocol. Balance tests included a one-legged stance time (OLST), which was timed for a maximum of 60 seconds, and functional reach using a yardstick mounted on the wall at shoulder height. Both of these tests have established reliability and are predictors of fall risk in older adults.22,23,29–32 In addition, the 10-Step Test was used to assess dynamic balance because it simulates the weight-shifting that occurs during stair ambulation. In the 10-Step Test, participants alternately placed one foot on top of a 10-cm step, 5 times per foot, to assess their dynamic balance when weight-shifting. This test was described by Miyamoto et al,33,34 who found it to be a reliable measure of agility and valid predictor of fall risk for older women. Gait speed was calculated by asking participants to ambulate on a level floor at their usual, self-selected pace. Using a stopwatch, the time it took each participant to cover a distance of 5 m (marked off by tape on the floor) during that walk was divided by 5 to calculate gait speed in m/s. Gait speed has been shown to be a reliable measure of functional mobility and a valid predictor of frailty and mortality in older adults.13,32,35–37
The various test sites used in this study did not permit standardization of the staircase used for the stair-climbing test; however, all of the staircases had a firm surface, handrails, and included either 8 or 10 steps. Using a stopwatch to record time, separate times were recorded in seconds for stair ascent and descent, then added together for combined stair-climbing time. To standardize our outcome measure, each time was divided by the number of steps to calculate the participants' stair-climbing speed in steps per second. Participants were allowed to use their assistive device and/or a handrail as needed during the stair-climbing test. This test protocol is similar to that used by other investigators who found it to have acceptable reliability (intraclass correlation = 0.84–0.86).38
Data Analysis Procedures
Descriptive statistics were used to calculate the average speed of stair ascent, descent, and combined ascent-descent for this sample of community-dwelling older adults. Stepwise regression techniques were used to analyze the best indicators of participants' combined ascent-descent speed (our dependent variable); separate analyses of ascent and descent speeds were also conducted. Independent variables in this analysis included the 6 functional mobility tests, as well as participants' age and body weight; the effect of gender was analyzed in a separate regression block. The underlying assumptions of normality, homoscedasticity, and linearity were analyzed by using a scatter plot of standardized residuals (Y) against standardized predicated values (X). Tolerance values were calculated for each independent variable in the regression model to determine what proportion of the variance in each variable was not accounted for by other variables in the model; tolerance values greater than 0.1 were considered acceptable. All data were analyzed by using PASW 18.0 statistical software (SPSS Inc., Chicago, IL) at a .05 α level. Given this α level and the number of variables included in our regression analysis, we projected a sample size of 50 to attain a statistical power of at least .80.
One outlier was identified and eliminated from the data set; thus, data from only 53 participants were analyzed. Descriptive data for participants' age, body weight, and functional test performances are presented in Table 1. Participants in this study ascended and descended stairs at an average speed of 1.3 (0.5) steps per second; ascent and descent speeds were similar. However, when analyzing stair-climbing speed by gender group, men demonstrated significantly faster speeds than women (1.6 vs 1.2 steps per second; t = 3.056; P = .004).
Results of the stepwise regression indicated that the 2 best indicators of stair-climbing speed were gait speed and OLST (R = 0.792; P < .001); these functional measures explained 63% of the variance in stair-climbing speed. Visual analysis of the scatter plot pattern of the residual and predicted values indicated that the regression model met the underlying assumptions of normality, homoscedasticity, and linearity. Figure 1 illustrates the moderate linear relationship between each of these predictors and stair-climbing speed. The tolerance values of these 2 variables were close to 1.0, indicating a high degree of independence between them. None of the remaining variables, including gender, that were excluded from the model were significantly associated with stair-climbing speed, although the P value for age was .057. Using the coefficients generated by the regression analysis (Table 2), the following explanatory equation was formulated for stair-climbing speed: steps per second = 2.335 − (gait speed × 1.156) + (OLST × 0.007). A paired t test was used to compare actual and predicted (equation) values of stair-climbing speed. The 2 values were not significantly different (t = −1.034; P = .306); mean values for actual and predicted speeds were nearly identical (1.32 vs 1.37 steps per second).
When analyzing the explanatory variables for ascent and descent speeds separately, gait speed and age were found to be the best indicators for ascent speed (R = 0.744; R2 = 0.553), while gait speed and OLST remained the best explanatory variables for descent speed (R = 0.774; R2 = 0.598). A bivariate analysis indicated that age was moderately correlated with both gait speed (r = 0.447) and OLST (r = −0.587).
Our results indicate that community-dwelling older adults ascend and descend an 8- to 10-step flight of stairs at an average pace of 1.3 steps per second. In a recent study by Oh-Park et al,39 the time needed to descend a flight of 3 stairs was found to be a significant predictor of functional decline among relatively high-functioning older adults who reported no difficulty in stair negotiation. When we converted their stair ascent and descent times to steps per second, their speeds were slightly less than those found in our study (∼1.1 steps per second) for a similar population. In another regression study similar to ours, Tiedemann et al38 reported an average stair ascent speed of 1.6 steps per second and descent speed of 1.7 steps per second. As was seen in our study, the men ambulated stairs slightly faster than the women. In their regression analysis, these investigators found that measures of knee extension and flexion strength, proprioception, edge contrast sensitivity, simple reaction time, dynamic balance, fear of falling, pain, and vitality significantly explained stair-climbing speed (R = 0.68 for ascent times; R = 0.71 for descent times) and explained 47% to 50% of the variance in these measures. Our analysis was able to explain a greater percent of the variance in stair-climbing speed on the basis of the results of only 2 functional tests that are quick and easy to administer. Thus, we believe our prediction model will be more useful to individuals who want to accurately predict an older adult's stair-climbing speed without much time, information, or skill. We were somewhat surprised that none of the more dynamic strength and balance tests such as timed chair stands, functional reach, or the 10-Step Test were included as independent variables in our regression model. However, like age, these measures were all moderately correlated with gait speed (P < .01), which remained the strongest single independent variable in the model.
Although our results were obtained from a group of older adults, similar findings were reported by Zaino et al,40 who investigated the reliability and validity of a Timed Up and Down Stairs Test among children with and without cerebral palsy. When we converted their Timed Up and Down Stairs Test time to speed scores, the children without cerebral palsy averaged 1.7 steps per second, while the children with cerebral palsy ranged from 0.6 to 0.9 steps per second, depending on the severity of their motor dysfunction. Interestingly, they also correlated the Timed Up and Down Stairs Test times with other functional measures and found those to be most highly correlated with the Timed Up-and-Go Test (rs = 0.78) and OLST (rs= 0.77) across all participants. Thus, the findings of our study, using a geriatric sample, were similar to this study, which focused on a pediatric population.
Even though our regression model accounted for more than half of the variance in stair-climbing speed, other factors such as living conditions and underlying pathologies were not considered in our analysis. Previous investigators have reported that factors such as living alone, visual deficits, arthritis, and heart disease can significantly impair one's ability to ambulate stairs.8,38,41 In a planned follow-up study, we will be conducting similar tests on a group of more frail older adults who are unable to climb stairs independently, to determine which tests best discriminate between those who can and cannot safely negotiate a flight of stairs.
Other factors that may have affected our test results include variations in the lighting and stairway configuration at each test site, along with variation in the participants' footwear. Although one stairway included a 90° turn, only the straight portion of the stairway was used for testing. All stairways had a firm surface, were of similar step height, and included handrails. Participants were allowed to use a handrail as a safety measure, if needed. However, we observed some participants using these handrails to pull themselves up the stairs, rather than simply using them for balance support. Because we did not record these observations, it was not considered in our data analysis. Nevertheless, other investigators have reported that participants who used handrails during stair ascent and descent had poorer vision, strength, and balance than those who ambulated the stairs without the use of handrails.38
The results of this study suggest that stair-climbing speed in older adults can be accurately estimated by using measures of his or her gait speed and OLST. Although it appears that a stair-climbing speed of 1.3 steps per second is normal for most healthy, community-dwelling older adults when ascending and descending an average-size staircase, further research is needed to determine whether this speed is sufficient for negotiating stairs in an emergency situation. In addition, further study is needed to determine which tests or measures best differentiate between individuals who can and cannot independently ascend and descend a flight of stairs. This information will help health care professionals and directors of residential communities make appropriate decisions related to the living accommodations needed for their older adult clients.
1. US Fire Administration. Fire Risks for the Older Adult. Emmitsburg, MD: US Fire Administration; 1999.
2. Gordon E, Said C, Galea M. Mobility on discharge from an aged care unit. Physiother Res Int. 2007;12(2):72–81.
3. Hausdorff JM, Rios DA, Edelber HK. Gait variability and fall risk in community-living older adults: a 1-year prospective study. Arch Phys Med Rehabil. 2001;82(8):1050–1056.
4. Hornbrook MC, Stevens VJ, Wingfield DJ, Hollis JF, Greenlick MR, Ory MG. Preventing falls among community-dwelling older persons: results from a randomized trial. The Gerontologist. 1994;34(1):16–23.
5. Lord SR, Ward JA, Williams P, Anstey KJ. An epidemiological study of falls in older community-dwelling women: the Randwick Falls and Fracture Study. Aust J Publ Health. 1993;17(3):240–245.
6. Hemenway D, Solnick SJ, Koeck C, Kytir J. The incidence of stairway injuries in Austria. Accid Anal Prev. 1994;26(5):675–679.
7. 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.
8. Startzell JK, Owens DA, Mulfinger LM, Cavanagh PR. Stair negotiation in older people: a review. J Am Geriatr Soc. 2000;48(5):567–580.
9. Wyatt JP, Beard D, Busuttil A. Fatal falls downstairs. Injury. 1999;30(1):31–34.
10. National Safety Council. NSC Injury Facts® 2012 Edition: A Complete Reference for Injury and Death Statistics. Itasca, IL: National Safety Council; 2012:20.
11. van Iersal MB, Rikkert MGM, Mulley GP. Is stair negotiation measured appropriately in functional assessment scales? Clin Rehabil. 2003;17(3):325–333.
12. Fried LP, Bandeen-Roche K, Williamson JD, et al. Functional decline in older adults: expanding methods of ascertainment. J Gerontol Med Sci. 1996;51A(5):206–214.
13. Bohannon RW, Brennan P, Pescatello L, Hasson S, Murphy M, Marschke L. Relationships between perceived limitations in stair climbing and lower limb strength, body mass index, and self-reported stair climbing activity. Top Geriatr Rehabil. 2005;21(4):350–355.
14. Cornette P, Swine C, Malhomme B, Gillet JB, Meert P, D'Hoore W. Early evaluation of the risk of functional decline following hospitalization of older patients: development of a predictive tool. Eur J Public Health. 2005;16(2):203–208.
15. Verghese J, Wang C, Xeu X, Holtzer R. Self-reported difficulty in climbing up or down stairs in nondisabled elderly. Arch Phys Med Rehabil. 2008;89(1):100–103.
16. Bean JF, Kiely DK, Herman S. The relationship between leg power and physical performance in mobility-limited older people. J Am Geriatr Soc. 2002;50(3):461–467.
17. Suzuki T, Bean JF, Fielding RA. Muscle power of the ankle plantarflexors predicts functional performance in community-dwelling older women. J Am Geriatr Soc. 2001;49(9):1161–1167.
18. Carmeli E, Reznick AZ, Coleman R, Carmeli V. Muscle strength and mass of lower extremities in relation to functional abilities in elderly adults. Gerontology. 2000;46(5):249–257.
19. Menz HP, Lord SR. The contribution of foot problems to mobility impairment and falls in community-dwelling older people. J Am Geriatr Soc. 2001;49(12):1651–1656.
20. Ried KF, Naumova EN, Carabello RJ, Phillips EM, Fielding RA. Lower extremity muscle mass predicts functional performance in mobility-limited elders. J Nutr Health Aging. 2008;12(7):493–498.
21. Curb JD, Ceria-Ulep CD, Rodriguez BL, et al. Performance-based measures of physical function for high-function populations. J Am Geriatr Soc. 2006;54(5):737–742.
22. Franchignoni F, Tesio L, Marino MT, Ricupero C. Reliability of four, simple quantitative tests of balance and mobility in healthy elderly females. Aging. 1998;10(1):26–31.
23. Kim M-J, Seino S, Kim M-K, et al. Validation of lower extremity performance tests for determining the mobility limitation levels in community-dwelling older women. Aging Clin Exp Res. 2009;21(6):437–444.
24. Mong Y, Teo TW, Ng SS. 5-repetition sit-to-stand test in subjects with chronic stroke: reliability and validity. Arch Phys Med Rehabil. 2010;91(3):407–413.
25. Ross MD, Fontenot EG. Test-retest reliability of the standing heel-rise test. J Sport Rehabil. 2000;9:117–123.
26. Schaubert KL, Bohannon RW. Reliability and validity of three strength measures obtained from community-dwelling elderly persons. J Strength Cond. 2005;19(3):717–720.
27. Segura-Ortí E, Martínez-Olmos FJ. Test-retest reliability and minimal detectable change scores for sit-to-stand-to-sit tests, the Six-Minute Walk Test, the One-Leg Heel-Rise Test, and handgrip strength in people undergoing hemodialysis. Phys Ther. 2011;91(8):1244–1252.
28. Yocum A, McCoy SW, Bjornson KF, Mullens P, Burton GN. Reliability and validity of the standing heel-rise test. Phys Occup Ther Pediatr. 2010;30(3):190–204.
29. Giorgetti MM, Harris BA, Jette A. Reliability of clinical balance outcome measures in the elderly. Physiother Res Int. 1998;3(4):274–283.
30. Lin M-R, Hwang H-F, Hu M-H, Wu H-DI, Wang Y-W, Huang F-C. Psychometric comparisons of the Timed Up and Go, One-Leg Stand, Functional Reach, and Tinetti Balance measures in community-dwelling older people. J Am Geriatr Soc. 2004;52:1343–1348.
31. Duncan PW, Studenski S, Chandler J, Prescott B. Functional reach: predictive validity in a sample of elderly male veterans. J Gerontol. 1992;47(3):M93–M98.
32. Sherrington C, Lord SR. Reliability of simple portable tests of physical performance in older people after hip fracture. Clin Rehabil. 2005;19:496–504.
33. Miyamoto K, Takebayashi H, Takimoto K, et al. The criterion-related validity of the ten step test compared with motor reaction time. J Phys Ther Sci. 2008;20:261–265.
34. Miyamoto K, Takebayashi H, Takimoto K, Miyamoto S, Morioka S, Yagi F. A new simple performance test focused on agility in elderly people: the ten step test. Gerontology. 2008;54(6):365–372.
35. Bohannon RW. Comfortable and maximum walking speed of adults aged 20-79 years: reference values and determinants. Age Ageing. 1997;26(1):15–19.
36. Dumurgier H, Elbaz A, Ducimetiere P, Tavernier B, Alperovitch A, Tzourio C. Slow walking speed and cardiovascular death in well functioning older adults: prospective cohort study. BMJ. 2009;339:b4460.
37. Harwood RH, Conroy SP. Slow walking speed in elderly people. BMJ. 2009;339:b4236.
38. Tiedemann AC, Sherrington C, Lord SR. Physical and psychological factors associated with stair negotiation performance in older people. J Gerontol A Biol Sci Med Sci. 2007;62(11):1259–1265.
39. Oh-Park M, Wang C, Verghese J. Stair negotiation time in community-dwelling older adults: normative values and association with functional decline. Arch Phys Med Rehabil. 2011;92(12):2006–2011.
40. Zaino CA, Marhese VG, Westcott SL. Timed Up and Down Stairs Test: preliminary reliability and validity of a new measure of functional mobility. Pediatr Phys Ther. 2004;16(2):90–98.
41. McLaughlin D, Leung J, Byles J, Dobson A. Living with stairs: functioning in a large cohort of older Australian adults. J Am Geriatr Soc. 2011;59(8):1560–1561.
functional predictors; gait speed; home safety; older adults; stair climbing
Copyright © 2014 the Section on Geriatrics of the American Physical Therapy Association
Highlight selected keywords in the article text.