This study is a single-blind, randomized, multi-center trial design. People who are admitted post-stroke for inpatient rehabilitation will be examined for eligibility by the treating physician. Subjects who satisfy the selection criteria and who agree to participate will be randomized to either the experimental group or the control group.
Intervention
All subjects will receive usual therapy during their inpatient rehabilitation stay.
Subjects in the experimental group (Daily Reinforcement of Speed or DRS group) will receive encouragement to walk faster with feedback on their performance every day for the duration of their inpatient stay. During one of the daily therapy sessions (preferably morning), the experimental subjects will be encouraged by their physical therapist to walk at least 10 m (25 feet) along a straight hallway. Subjects in the experimental group will be timed with a stopwatch by their treating physical therapist each day during the 10-meter walking task. If a subject is not able to walk 10 meters, then a shorter distance is used until the 10-meter distance can be achieved (if possible). Before starting the timed walk, the therapist will remind the subject to "walk as fast as safely possible" and to try to walk the distance "a few seconds faster than what you were able to do in the past few days". The therapist can provide the assistance of another person and use aids, and whatever is needed for safety. The timed test should be stopped or not performed on any one day if this is best for patient care. The subject's safety must not be compromised. At the end of the timed walk, the therapist will provide the results for that day in a positive tone, such as "Very good! You walked that in xx seconds". Then, (a) 'This is better by x seconds', or (b) 'This shows you are holding your own', or (c) 'I believe that you will soon be able to walk a bit faster with a better walking pattern'."
Subjects who are randomized to the control group (No Reinforcement of Speed or NRS group) will not be timed during a daily 10-meter walk, nor specifically encouraged to walk faster each day.
Outcome measures
The primary outcome measure is walking speed, measured with a stopwatch over 15.2 m (50 feet).3 A blinded evaluator (i.e. a physical therapist or other trained person who is not involved in the subject's treatment and does not work on the inpatient unit) will conduct all assessments. The treating physical therapist will assist the subject during the gait speed assessments as necessary. The subject will employ his or her usual assistive devices for that stage of recovery.
The gait speed assessment will be conducted on a flat tiled surface along a walkway that is not within sight of other patients. A chair will be placed at either end of the 15 m straight walkway. The subject starts approximately 3 steps behind the start line to allow for acceleration. The blinded assessor instructs the subject to "walk as fast as you safely can until I tell you to stop. If you have to sit or rest, tell me." When the subject's lead toe crosses the line, timing begins. When the lead toe crosses the 15.2-meter (finish) line, timing stop. The subject is told to stop walking once both feet cross the line. This ensures that the subject does not decelerate until he or she has traversed the full 15.2 meters. Data is recorded as time in seconds.
At entry to the study, it is anticipated that some subjects may not be able to walk continuously for 15 meters. In these cases, a timed 5-meter walk can be substituted at study entry and designated so on the data entry form, and if still necessary, at discharge. If a subject at study entry cannot walk 5 meters without stopping to rest or needs maximal assistance, the test is marked "unable" on the data recording sheet.
Gait speed measurements will be collected by the blinded evaluator at study entry (if possible), at 1, 2, 4, 6, and 8 weeks after entry if still an inpatient, within one day of inpatient discharge, and, if possible, 3 months and 6 months after study entry. The Functional Ambulation Classification Index will be completed by the blinded evaluator after each gait speed assessment. Thus, a FAC score will be recorded at the same time points as each gait speed evaluation.
The secondary outcome measure is distance walked in 3 minutes. The subject will be instructed by the blinded assessor to walk as many laps of the walkway as possible in 3 minutes. A straight 25-meter walkway will be used. The patient turns at each end and continues walking for the 3 minutes. The distance covered in 3 minutes is measured in meters. The subject may halt to rest at any time and continue walking when ready. If the subject must sit down to rest, the test is discontinued and the time and distance are recorded. The 3-minute walking test will be conducted at up to 4 time points: 4 weeks after entry if still an inpatient, within 2 days of inpatient discharge, 3 months post study entry, and 6 months post study entry. If the evaluator believes that the test was not an accurate reflection of the walking distance (some interruption occurred or the subject stopped when distracted), the test can be repeated and the best distance recorded.
The blinded observer should find an isolated and convenient place to perform the tests. Before the trial begins, the observer should practice the timed walking speed and distance test with 3 hemiparetic inpatients and their therapists for 4 repeated tests within 1-2 days to try to obtain reproducible data. The walking speeds and distances should not differ by more than 10%.
Descriptive data and further assessments
Each subject will have data recorded by site and randomization number, not by name, birth date or other identifying characteristics. In addition to the results from the gait speed and walking distance assessments (collected and entered by the blinded evaluator), basic descriptive information for each subject will also be collected in order to adequately characterize the study population. This information will be obtained by the study physician or designated person from the medical record and from interview with the subject or family. The following information will be collected for each subject: age, gender, handedness; dates of stroke onset, admission for rehabilitation, entry into the trial, and discharge from in patient care; and (selected from a list) type of stroke, general location of stroke, and side of hemiparesis. Severity of stroke at the time of entry into the study will be characterized using the National Institutes of Health Stroke Scale,4-5 the modified Rankin Scale, and the Functional Ambulation Classification Index, which are well accepted measures. The investigators will also ask subjects to record the number of falls they had from discharge to 3 months and from 3 months to 6 months after entry using a calendar that they will bring with them to each follow-up.
Data entry
All data entry will be via a web-based system. The blinded observer will enter all walking-related data and will be given a password to access the data entry site that differs from the password and access by the study physician.
Data analysis
For an anticipated effect size of 0.4, a conservative estimate using a two-tailed alpha of 0.01 and power of 0.9, we will need a sample size of 150 subjects in each group. An ideal sample size for this varied population and for the primary and secondary outcomes will be 250 subjects in each arm to allow for dropouts and to gather enough subjects who are also tested at 3 and 6 months study entry. This would make the trial the largest prospective neurorehabilitation study ever attempted.
