Share this article on:

Gait Speed Is Limited but Improves Over the Course of Acute Care Physical Therapy

Braden, Heather J. PT, MPT, PhD, GCS; Hilgenberg, Sean DPT; Bohannon, Richard W. DPT, EdD; Ko, Man-Soo PhD; Hasson, Scott PT, EdD, FACSM, FAPTA

Journal of Geriatric Physical Therapy: July/September 2012 - Volume 35 - Issue 3 - p 140–144
doi: 10.1519/JPT.0b013e31824baa1e
Research Reports

Background: Gait is a common focus of physical therapists' management of patients in acute care settings. Walking speed, the distance a patient covers per unit time, has been advocated as a “sixth vital sign.” However, the feasibility of measuring walking speed and the degree to which walking speed is limited or improves over the course of therapy in the acute care setting are unclear.

Purpose: The purpose of this study of patients undergoing physical therapy during acute care hospitalization, therefore, was to determine whether walking speed can be measured in acute care and whether walking speed is limited and changes over the course of therapy.

Methods: This was an observational cross-sectional study. Participants were 46 hospital inpatients, mean age 75.0 years (SD = 7.8), referred to physical therapy and able to walk at least 20 ft. Information regarding diagnosis, comorbidities, physical assistance, device use, body height, and weight was obtained. Speed was determined during initial and final physical therapy visits while patients walked at their self-selected speed over a marked course in a hospital corridor.

Results: Therapists reported that walking speed was clinically feasible, requiring inexpensive, available resources, 4 minutes' additional time, and simple calculations for documentation. Initial walking speed was a mean of 0.33 m/s (SD = 0.21; 95% confidence interval [CI]: 0.27–0.39), whereas final speed was 0.37 m/s (SD = 0.20; 95% CI: 0.31–0.43). The Wilcoxon test showed the increase in walking speed (0.04 m/s) to be significant (P = .005) over a mean therapy period of 2.0 days (SD = 1.4) and total hospitalization period of 5.5 days (SD = 3.6). The effect size and standardized response mean were 0.19 and 0.36, respectively. Minimal detectable change was 0.18 m/s.

Conclusion: Walking speed is a feasible measure for patients admitted to an acute care hospital. It shows that patients walk slowly relative to community requirements but that their speed improves even over a short course of therapy.

1Physical Therapy Program, Department of Nursing and Rehabilitation Sciences, Angelo State University, San Angelo, Texas

2Rehabilitation Department, Shannon Medical Center, San Angelo, Texas

3Physical Therapy Program, Department of Kinesiology, University of Connecticut, Storrs.

Address correspondence to: Heather J Braden, PT, MPT, PhD, GCS, Physical Therapy Program, Department of Nursing and Rehabilitation Sciences, Angelo State University, San Angelo, TX. (

The authors declare no conflict of interest.

Back to Top | Article Outline


The ability to walk is an important focus of the physical therapy management for many patients. Walking speed is a principal element of ambulation recommended as a vital sign.1,2 It predicts health status such as hospital length of stay, medical costs in facilities, potential for rehabilitation,3 and mortality.4 It also explains walking behavior,5 fear of falling,6 functional decline,7 and home8 and community mobility.811 Measurements of walking speed have been shown to be reliable, valid, and responsive. Moreover, normative values are available,9,1215 however, feasibility and usefulness of measuring walking speed in acute care have received little attention. In fact, there is little evidence of its use in acute care settings.13,14 The medical record review completed by Bohannon and Sandstrom13 did not find a single note in which therapists documented walking speed. We, therefore, sought to determine whether walking speed can be measured in acute care, establishing feasibility, and whether it is useful for establishing patient status in acute care where stays are increasingly short.15

Reported barriers to using outcome measures for walking ability include lack of consensus about which measure to use, time to test and properly document findings, and availability of measures relevant to home and community environments.16 Walking speed is a meaningful test that takes only minutes to complete and uses readily available resources such as a stopwatch and a facility hallway.17 Although gait assessment and gait training are commonly observed in acute care patients, it is unclear how much walking speed is limited and whether it improves over the course of therapy. Identifying trends for walking speed may be useful for selecting patients in need of ongoing therapy either in the current setting or in the next level of care.

The purpose of this study of patients referred to physical therapy in an acute care setting was (1) to evaluate whether walking speed can be measured and (2) to describe typical walking speed at the beginning and end of a physical therapy intervention.

Back to Top | Article Outline


This was a pre-post observational cross-sectional study approved by the Angelo State University and Shannon Medical Center institutional review boards. This study involved patients admitted to the acute care setting with physical therapy orders. The acute care hospital consisted of a total of 200 beds; referral for physical therapy was made for 76% of patients older than 60.0 years. Patients typically were seen for one 30-minute session per day; interventions included bed mobility and transfer practice at bedside, followed by ambulation training in the hallway. Consenting patients were examined on their initial and final (predischarge) physical therapy visit.

Back to Top | Article Outline


This study aimed to investigate walking speed among hospitalized older adults. Potential participants had to be at least 60.0 years of age, and able to walk a distance of at least 20 ft. Potential participants were excluded if they did not have ambulatory potential, if they had contraindications to ambulation, or if they were younger than 60.0 years. Patients who did not have ambulatory potential included those whose prior baseline level of function was wheelchair-level mobility. Seventy patients referred to physical therapy were screened for the study over a period of 3 months. Sixteen were excluded because they were unable to complete the minimum walking distance, and 8 were excluded because they were younger than 60.0 years. The final sample size was 46 patients (65.7% of referrals in the 3-month study period).

Back to Top | Article Outline

Testing Procedure

Five physical therapists used a standardized procedure to measure walking speed for patients on their individual caseloads. The therapists were asked to provide feedback about how informative they found walking speed to be as well as their thoughts on the feasibility (ease of administration, time efficiency) of measuring walking speed in the acute care setting. This information was gathered via informal discussion including all 5 physical therapists with an opportunity to provide written, anonymous feedback if elaboration was needed. Walking speed was assessed during initial and final physical therapy visits. Patients walked at their self-selected speed over a marked course in a hospital corridor. Self-selected walking speed measures are comparable across varying distances.8,10 The 20-ft course included a 5-ft zone for acceleration, a central 10-ft “steady-state” zone that was timed with a stopwatch, and a 5-ft zone for deceleration. Walking speed has been measured as a single trial18,19 or by averaging multiple trials.8,20 A single trial of self-selected walking speed at initial and discharge examinations was utilized. Level of assistance (moderate, minimal, contact guard assist, supervision, or none) and device used (rolling walker, single-point cane, crutches, or none) were documented. Patient height and mass were documented. Patients were unaware of the timed start and stop points and received no feedback or training during trials.

Back to Top | Article Outline

Statistical Analysis

Clinical feasibility for using walking speed as an assessment tool in acute care was obtained via comments from the 5 physical therapists collecting data. Via informal discussion, therapists commented on whether measurement of walking speed was difficult to implement and how much time walking speed testing added to the physical therapy examination. All analyses were conducted via SPSS 18.0 (IBM Corporation, Armonk, New York). Descriptive statistics, effect size, and standardized response mean were calculated. Because of a small sample with walking speeds that were not normally distributed, the nonparametric Wilcoxon test was used to compare walking speed measurements between the initial and final therapy sessions.

Effect size was calculated by dividing the mean change between admission and discharge speed by the standard deviation (SD) of the admission speed. Standardized response mean was calculated by dividing the mean change between admission and discharge speed by the SD of the change between admission speed and discharge speed. Minimal detectable change (MDC95%) was calculated on the basis of an assumed intraclass correlation coefficient (ICC) of 0.90 as reported by Mangione et al.21 The value of ICC was then used to determine the standard error of measurement (SEM = SDadmission_WS × √1 − ICC), where WS is walking speed, and the MDC95% (SEM × 1.96 × √2).

Back to Top | Article Outline


The physical therapists who collected walking speed data during the study agreed that documenting walking speed provides an opportunity to show progress from beginning to end of physical therapy in acute care and to assist with discharge recommendations. All stated that the measurement of walking speed was not difficult to implement and should be considered to be one component of gait analysis completed during the physical therapist's examination of patients in acute care. The physical therapists reported that an extra 4 minutes was required to obtain patient consent, assess walking speed, and document the walking speed electronically within the typical 30-minute physical therapy session. Remembering to conduct the actual walking speed test as part of their initial examination process was reported to be the only challenge.

The average age of the participants was 75.0 years (SD = 7.8), with a range from 60.0 to 90.0 years. Thirty-one participants (67%) were women. The most common reasons for the acute care stay are listed in Table 1. Of those with orthopedic primary diagnoses, 11 of 18 had total joint replacements of the knee or hip and 1 had a hip fracture. Those with total knee or hip replacements received initial therapy intervention the day following surgery. Eighteen of the 46 subjects (39%) had 5 or more comorbidities, 12 (26%) had 3 or 4 comorbidities, 12 (26%) had 1 or 2 comorbidities, and 4 (9%) had no comorbidity. None of the participants were reported to have a fall while in acute care. Body mass index (BMI) ranged from 16.0 to 50.0 kg/m, with 13 patients (28%) classified as obese having 30.0 kg/m or more BMI. Forty-four of 46 patients used an assistive device. Forty-two of 46 required therapist assistance (Table 1). The mean duration of physical therapy was 2.0 days (SD = 1.4); mean length of hospital stays was 5.5 days (SD = 3.6).

Mean walking speed on admission was 0.33 m/s (SD = 0.21; CI: 0.27–0.39) (Table 2). Mean walking speed on discharge was 0.37 m/s (SD = 0.20; CI: 0.31–0.43). The speed of 19 participants improved, for 19 stayed the same, and for 8 declined (Figure). The increase in walking speed from admission to discharge (0.04 m/s, SD = 0.11; CI: 0.01–0.07) was significant (P = .005). The effect size and standardized response mean were 0.19 and 0.36, respectively. Both were small according to Cohen criteria. On the basis of an ICC of 0.90, the associated MDC95% for walking speed was 0.18 m/s.

Back to Top | Article Outline


Feasibility results in this study indicate that walking speed can indeed be measured in acute care and encourage the use of walking speed as a practical assessment tool in this setting. If walking speed can be successfully measured in acute care settings as a standard examination tool, the results may assist health care professionals with readiness to discharge from the hospital and proper placement into a safe environment, thus preventing potential future injury or illness and escalating health care costs.

This study reports change in walking speed from start to end of physical therapy care for a diverse sample of patients in a general acute care setting. Patients in this study walked at less than half of the typical walking speeds of healthy community-living older adults. Typical walking speed has been reported as 1.34 m/s for 60.0- to 69.0-year-old men and 1.24 m/s for 60.0- to 69.0-year-old women.22 For this study, patients in this age range walked at a mean speed on discharge of 0.54 m/s for men and 0.40 m/s for women. Community-living men between 70.0 to 79.0 years typically walk at a speed of 1.26 m/s and women at 1.13 m/s. Walking speed on discharge among patients in this study for 70.0- to 79.0-year-olds was 0.42 m/s for men and 0.24 m/s for women.22 Community-living men aged 80.0 to 99.0 years typically walk at a speed of 0.97 m/s, whereas women 80.0 to 99.0 years typically walk at a speed of 0.94 m/s.22 The oldest male patients in this study walked at a speed of 0.45 m/s, whereas the oldest women walked at a speed of 0.30 m/s.

Although patients were slower than their community-dwelling peers, they demonstrated meaningful improvement in walking speed over a short course of therapy. Such improvements provide support for administering this assessment tool in the busy acute care setting. The degree of change in walking speed may be related to both primary diagnosis and overall burden of acute and chronic illness. The effect size (0.19) falls in the lower range of 0.10 to 0.83 reported by others.2326 This low effect size may be related to the short length of stay typical of the acute care environment. The standardized response mean (0.36) is less than the range of 0.88 to 1.10 reported by others2629 and also likely due to a shorter length of stay by the participants of the current study. The MDC95% (0.18 m/s) calculated from walking speed data collected in this study is similar to that of others21 and might be useful in documentation and for clinical interpretation.

The ability of walking speed to predict patient function and health care needs is supported in clinical research literature. In an epidemiologic study of frail, older male veterans, each 0.10-m/s decrease in baseline walking speed indicated poorer health status, likelihood of longer hospital stays, and higher medical costs ($1334 per year).17 Each 0.10-m/s increase in walking speed predicted likelihood of improved physical function, fewer hospitalization days (2.3), and 1-year medical cost reductions of $1188.17 Walking speeds less than 0.15 m/s predict likelihood of discharge to skilled nursing or inpatient rehabilitation facilities rather than to home.30

Overall hospitalization for the current study averaged 5.5 days (SD = 3.6) for patients, and therapy duration averaged only 2.0 days (SD = 1.4). Future studies may consider whether receiving physical therapy for more days during total hospitalization might better prepare patients for community reentry prior to discharge and result in cost savings in health care. Future studies could include a control group to identify walking speed improvements in acute care patients who had no therapy orders just as general health improved during hospitalization. Body mass index was recorded to observe whether the presence or absence of obesity might impact walking speed. Obese older adults have a greater likelihood of discharging to nursing homes due to the tendency for accelerated functional decline that accompanies obesity.2 Obese older adults have a normal walking speed of 1.23 m/s,2 and obese participants in the current study averaged 0.37 m/s upon therapy discharge. With regard to community standards, acute care patients at therapy discharge were on average 0.12 m/s slower than the velocity required to cross the street at a traffic signal (0.49 m/s).31 Although walking speed improvements were statistically significant from initial therapy to discharge of therapy and the trend toward improved walking speed is apparent during therapy in acute care, patients on average failed to return to community-level ambulation after their acute care stay. Only 9% of patients in this study increased walking speed beyond the estimated MDC95%. Future studies will consider actual discharge destination from acute care, falls that occur after discharge, and rehospitalizations with respect to the patient's discharge walking velocity. Future studies will also investigate whether continued improvements occur with walking speed at the next level of care, such as skilled nursing facilities or inpatient rehabilitation facilities, where lengths of stay are greater.

This study had several limitations. First, the sample size was limited. Although this limitation did not preclude the finding of significant improvements in walking speed over the course of physical therapy, it might affect the estimates of responsiveness. Second, the study included the substantial number of patients with lower-extremity orthopedic issues. Although this may be typical in such settings, it may result in findings not indicative of other diagnostic groups. Third, this study involved a single site. Whether therapists in other sites would agree regarding the feasibility of measuring walking speed requires further investigation. Also warranting investigation in other acute settings are the limitations and changes in walking speed identified herein.

Back to Top | Article Outline


Walking speed is a feasible and informative measure for patients admitted to an acute care hospital. It revealed limitations among the patients relative to normal community-dwelling adults. This study also revealed that the patients' walking improved even over a short course of therapy. With patients being released from acute care hospitalization sooner, consideration for safe discharge location is imperative.

Back to Top | Article Outline


The authors thank Dr Carolyn Mason for encouraging the vision of this study and Kaitlin O'Halloran, SPT, and Maddie Bohmfalk, SPT, for their motivation and dedication to this project.

Back to Top | Article Outline


1. Fritz S, Lusardi M. White paper: walking speed the sixth vital sign. J Geriatr Phys Ther. 2009;32:46–49.
2. Villareal DTBM, Siener C, Sinacore DR, Klein S Physical frailty and body composition in obese elderly men and women. Obes Res. 2004;12:913–920.
3. Goldie PA, Matyas TA, Evans OM. Deficit and change in gait velocity during rehabilitation after stroke. Arch Phys Med Rehabil. 1996;77:1074–1082.
4. Hardy SE, Perera S, Roumani YF, Chandler JM, Studenski SA. Improvement in usual gait speed predicts better survival in older adults. J Am Geriatr Soc. 2007;55:1727–1734.
5. Bowden MG, Balasubramanian CK, Behrman AL, Kautz SA. Validation of a speed-based classification system using quantitative measures of walking performance poststroke. Neurorehabil Neural Repair. 2008;22:672–675.
6. Maki BE. Gait changes in older adults: predictors of falls or indicators of fear. J Am Geriatr Soc. 1997;45:313–320.
7. Montero-Odasso M, Schapira M, Soriano ER, et al. Gait velocity as a single predictor of adverse events in healthy seniors aged 75 years and older. J Gerontol A Biol Sci Med Sci. 2005;60:1304–1309.
8. Bohannon RW. Measurement of gait speed of older adults is feasible and informative in a home-care setting. J Geriatr Phys Ther. 2009;32:22–23.
9. Vos-Vromans DC, de Bie RA, Erdmann PG, van Meeteren NU. The responsiveness of the ten-meter walking test and other measures in patients with hemiparesis in the acute phase. Physiother Theory Pract. 2005;21:173–180.
10. Bohannon RW. Population representative gait speed and its determinants. J Geriatr Phys Ther. 2008;31:49–52.
11. Langlois JA, Keyl PM, Guralnik JM, Foley DJ, Marottoli RA, Wallace RB. Characteristics of older pedestrians who have difficulty crossing the street. Am J Public Health. 1997;87:393–397.
12. Barthuly AM, Bohannon RW, Gorack W. Gait speed is a responsive measure of physical performance for patients undergoing short-term rehabilitation. Gait Posture. 2011. In Press.
13. Bohannon R, Sandstrom RN. Documentation of ambulation in initial therapy notes. Acute Care Perspect. 2003;13:14–15.
14. Bohannon RW, Ahlquist M. Documentation of prestroke ambulation. Int J Rehabil Res. 2003;26:71–72.
15. Directorate for Employment Labour, and Social Affairs. OECD Health Data 2011.,2340,en_2649_34631_2085200_1_1_1_1,00.html. Published 2011. Accessed October 28, 2011.
16. Salbach N, Guilcher SJ, Jaglal SB. Physical therapists' perceptions and use of standardized assessments of walking ability post-stroke. J Rehabil Med. 2011;43:543–549.
17. Purser JL, Weinberger M, Cohen HJ, et al. Walking speed predicts health status and hospital costs for frail elderly male veterans. J Rehabil Res Dev. 2005;42:535–546.
18. Graham JEOG, Fisher SR, Ottenbacher KJ. Assessing walking speed in clinical research: a systematic review. J Eval Clin Pract. 2008;14:552–562.
19. Graham JEFS, Berges IM, Kuo YF, Ostir GV. Walking speed threshold for classifying walking independence in hospitalized older adults. Phys Ther. 2010;90:1591–1597.
20. Lusardi MPG, Schulman M. Functional performance in community living older adults. J Geriatr Phys Ther. 2003;26:14–22.
21. Mangione KKCR, McCormick AA, Blevins HL, White MB, Sullivan-Marx EM, Tomlinson JD. Detectable changes in physical performance measures in elderly African Americans. Phys Ther. 2010;90:921–927.
22. Bohannon RW, Williams Andrews A. Normal walking speed: a descriptive meta-analysis. Physiotherapy. 2011;97:182–189.
23. Conklyn D, Stough D, Novak E, Paczak S, Chemali K, Bethoux F. A home-based walking program using rhythmic auditory stimulation improves gait performance in patients with multiple sclerosis: a pilot study. Neurorehabil Neural Repair. 2010;24:835–842.
24. English CK, Hillier SL, Stiller K, Warden-Flood A. The sensitivity of three commonly used outcome measures to detect change amongst patients receiving inpatient rehabilitation following stroke. Clin Rehabil. 2006;20:52–55.
25. Salbach NM, Mayo NE, Higgins J, Ahmed S, Finch LE, Richards CL. Responsiveness and predictability of gait speed and other disability measures in acute stroke. Arch Phys Med Rehabil. 2001;82:1204–1212.
26. Van Lersel M, Munneke M, Esselink RA, Benraad CE, Olde Rikkert MG. Gait velocity and the Timed-Up-and-Go test were sensitive to changes in mobility in frail elderly patients. J Clin Epidemiol. 2008;61:186–191.
27. Ahmed S, Mayo NE, Higgins J, Salbach NM, Finch L, Wood-Dauphinee SL. The Stroke Rehabilitation Assessment of Movement (STREAM): a comparison with other measures used to evaluate effects of stroke and rehabilitation. Phys Ther. 2003;83:617–630.
28. Garland SJ, Willems DA, Ivanova TD, Miller KJ. Recovery of standing balance and functional mobility after stroke. Arch Phys Med Rehabil. 2003;84:1753–1759.
29. Francisco GE, Boake C. Improvement in walking speed in poststroke spastic hemiplegia after intrathecal baclofen therapy: a preliminary study. Arch Phys Med Rehabil. 2003;84:1194–1199.
30. Rabadi MHBA. Admission ambulation velocity predicts length of stay and discharge disposition following stroke in an acute rehabilitation hospital. Neurorehabil Neural Repair. 2005;19:20–26.
31. Andrews AWCS, Bourassa M, Garvin M, Benton D, Tanner S. Update on distance and velocity requirements for community ambulation. J Geriatr Phys Ther. 2010;33:128–134.

acute care; geriatric assessment; mobility; walking speed

© 2012 Academy of Geriatric Physical Therapy, APTA