Cognitive Function is a Predictor of the Daily Step Count in Patients With Subacute Stroke With Independent Walking Ability: A Prospective Cohort Study

Highlights • Cognition at admission may predict daily step count.• Cognitive impairment may increase risk of poor ambulation after subacute stroke.• Ambulation poststroke is influenced by both physical and cognitive factors.

the MMSE score at admission (reference, 28-30 points; B, À2.07; 95% confidence interval, À3.89 to À0. 35; b, À0.22; P=.027) was significantly associated with the daily step count at discharge. Conclusions: Cognitive function at admission had a significant association with the daily step count at discharge in patients with subacute stroke who could walk independently. © 2021 The Authors. Published by Elsevier Inc. on behalf of American Congress of Rehabilitation Medicine. This is an open access article under the CC BY license (http://creativecommons.org/ licenses/by/4.0/). Stroke is a leading cause of disability, 1,2 and poststroke disability improves after rehabilitation. 3,4 Exercise-based therapy is widely known to enhance motor recovery after stroke, with evidence of the positive relationship between physical activity and recovery in a meta-analysis. 5 Moreover, physical activity in patients with stroke has been reported to be associated with the prevention of stroke recurrence, 6,7 prevention of readmission, 8 and improvement of physical health 1year poststroke. 9 Thus, it is recommended to perform physical activity during the rehabilitation period. 10 Although patients with stroke in the subacute phase undergo intensive rehabilitation to improve activities of daily living, with the expectation that it will help speed up recovery, 11 a systematic review reported that patients with subacute stroke are often inactive. 12 The number of steps taken by patients with subacute stroke is fewer than those taken by healthy adults 12,13 and patients with chronic stroke. 14 Furthermore, daily steps during hospitalization are related to those after discharge. 15,16 Therefore, established exercise habits are important, especially in the subacute rehabilitation hospital.
Very few reports have investigated the various factors associated with the daily step count in patients with subacute stroke. To our knowledge, only physical aspects such as walking speed 17 and physical function 18 are reportedly associated with the daily steps taken by patients with subacute stroke. We hypothesized that cognitive and psychological aspects, such as mild cognitive decline and motivation decline, also affect daily steps. Exploring various factors related to daily steps, including physical, cognitive, and psychological aspects, is important for establishing exercise habits in the subacute setting. In particular, patients with mild stroke who can walk independently usually have less support after home discharge; thus, it is necessary to investigate factors related to inactivity. However, the relationship between daily steps and physical, cognitive, and psychological aspects remains unclear in patients with subacute stroke with independent walking ability. Therefore, this study aimed to explore the factors associated with the daily step count in patients with subacute stroke who could walk independently, including physical, cognitive, and psychological factors.

Study design and participants
This prospective cohort study was conducted in adherence to the Strengthening the Reporting of Observational Studies in Epidemiology statement. This study included patients who were admitted to subacute rehabilitation wards between January 1, 2018 and July 31, 2019. The inclusion criteria were aged ≥20 years and admission with a first-ever stroke. The exclusion criteria included subarachnoid hemorrhage, cognitive impairment (Mini-Mental State Examination [MMSE] score ≤23 points), aphasia, early discharge (within 7 days of admission), hospital transfer, refusal to participate, and nonindependent walking during hospitalization. This study was conducted in accordance with the Declaration of Helsinki, and this study was reviewed and approved by the Ethics Committee of Tokyo Bay Rehabilitation Hospital (#189). All participants provided written informed consent before data collection.

Data collection
The following demographic data were collected from the medical records: age (y), sex (men or women), body mass index (kg/m 2 ), serum albumin level (g/dL), type of stroke (cerebral infarction or cerebral hemorrhage), side of the brain affected (right or left), duration from the stroke onset to admission (d), duration of hospitalization (d), circumstances of living (alone or not), discharge disposition (home or facility), and presence of neglect (yes or no). Data regarding duration of hospitalization and condition at discharge were collected at discharge, whereas the other data were collected at admission. The measurement scores recorded at admission and discharge are presented below. Assessments were completed within 1 week of admission and discharge.

Ambulatory activity
Ambulatory activity was operationally defined as the daily step count measured using a pedometer. In previous studies, ambulatory activity was assessed using a pedometer in patients with stroke, and this measurement method had an established feasibility, 19-21 validity, 22,23 and reliability. 23 We used a pedometer a with a 14-day data storage capacity that had a triaxial acceleration sensor that could measure daily steps. The pedometer was placed in the pants pocket on the nonparalyzed side. This pedometer has not been validated in patients with stroke; however, a previous study reported on the validity of the same measurement method as this study. 22 Patients were instructed to wear the pedometer during the day continuously for 7 consecutive days except while bathing and sleeping. The average daily steps tracked for at least 3 days was used for each patient in this study. 24 The primary outcome measure was the average daily steps at discharge.

Walking speed
Walking speed was assessed using a 10-m walking test, which has established reliability in patients with stroke. 25 Participants were instructed to walk at a comfortable speed, and the time required to walk from the starting line to the goal line was measured using a stopwatch. Walking speed was calculated by dividing the 10-m distance by the time required (m/s).

Stroke impairment assessment set
Motor and sensory functions were assessed using the stroke impairment assessment set. 26 Motor scores consist of 2 tests for upper extremity (0-10) and 3 tests for the lower extremity (0-15). 27 Sensory scores evaluate superficial sensation and deep sensation of the affected upper (0-6) and lower (0-6) extremities. 27 Higher scores represent better functions.

FIM
Functional disability was assessed using the FIM. 28 The FIM comprises 13 motor subscales (FIM motor) and 5 cognitive subscales (FIM cognitive). The FIM motor consists of the following 4 categories: self-care (eating, grooming, bathing, dressing-upper body, dressing-lower body, toileting), sphincter control (bladder management and bowel management), transfers (bed/chair/wheelchair, toilet, tub/shower), and locomotion (walk/wheelchair and stairs). The FIM cognitive consists of the following 2 categories: communication (comprehension and expression) and social cognition (social interaction, problem solving, memory). Each item has a 7grade scale ranging from 1 point (total assistance or not testable) to 7 points (complete independence). The total possible score is 18-126 points, 13-91 points, and 5-35 points for the total FIM, FIM motor, and FIM cognitive, respectively, with a higher score representing greater functional independence. The severity of function disability at discharge was categorized as mild (FIM motor≥62 points), moderate (FIM motor=38-61 points), or severe (FIM motor≤37 points) based on a previous study. 29

Self-Rating Depression Scale
Depressive symptoms were assessed using the Self-Rating Depression Scale, consisting of 20 items. 34 Each question is scored on the following 4-point scale: 1 point, rarely; 2 points, sometimes; 3 points, commonly; and 4 points, most of the time. Total possible scores are 20-80 points, with a higher score indicating more depressive symptoms; the cutoff value is 50 points. 35 The validity of the Self-Rating Depression Scale has been established in patients with stroke. 36

Apathy Scale
Motivation was assessed using the Apathy Scale, 37 which consists of 14 items. Each item is scored on the following 4point scale: 0 points, not at all; 1 point, slightly; 2 points, some; and 3 points, commonly. Total possible scores are 0-42 points, with a higher score indicating more apathy symptoms; the cutoff value is 16 points. 37 The Apathy Scale has established validity and reliability in patients with stroke. 38

Statistical analysis
The normality of the data was plotted using histograms and assessed using the Shapiro-Wilk test. To assess the factors associated with the daily step count at discharge, we used multiple regression analysis to determine the partial regression coefficients (B), 95% confidence interval [CI], standard partial regression coefficients (b), and variance inflation rate. Daily step count at discharge was the dependent variable. Daily steps at discharge were divided by 1000 to prevent the partial regression coefficient from becoming too large. 6 Independent variables were the factors with a P≤.05 in the univariate regression analysis. Independent variables without normality were categorized by cutoff values. All statistical analyses were performed using SPSS Statistics 21.0. b Values of P≤.05 were considered statistically significant.

Discussion
We investigated the factors associated with the daily step count at discharge in patients with subacute stroke with independent walking ability. In the multiple regression analysis, we demonstrated that the MMSE score at admission was mildly but significantly associated with the daily step ambulated at discharge. We found that the MMSE score at admission was related with the daily steps in multiple regression analysis. To our knowledge, there have been no studies that have investigated the relationship between cognitive function and ambulatory activity in patients with subacute stroke. For patients with chronic stroke, however, cognitive impairment is a risk factor for inactivity. 39 In the present study, we excluded patients with MMSE scores ≤23 points because patients with severe cognitive impairment were omitted. Our results suggested that even MCI affected ambulatory activity in patients with subacute stroke. According to the available data, this is the first study to demonstrate that the MCI at admission was associated with the daily step count at discharge in patients with subacute stroke. It has been reported that MCI was associated with low physical activity in older adults. [40][41][42] Similarly, in patients with stroke, MCI may interfere with active participation in rehabilitation and voluntary practices. Thus, cognitive screening tests to identify MCI and approaches to increase their activity may be important.
One strength of this study is that we identified factors associated with the daily step count in patients with subacute stroke in the multiple regression analysis, including cognitive aspects such as cognitive function, motivation, and depressive symptoms. Previous studies have reported that daily step count was related to only physical aspects 17,18 and did not clarify the influence of confounding factors on daily step count. Thus, our results after the adjustment for physical and cognitive confounding factors were novel.

Study limitations
There are some limitations to this study. First, the study was conducted at a single facility, which limits the generalizability of the results. Second, we used the psychological scale; thus, we excluded patients with a MMSE score ≤23 points. Finally, we could not measure activity time, content, or intensity because we used only a pedometer to measure daily steps. Nevertheless, the pedometer was easy to operate; hence, many patients with stroke agreed to participate in this study. Furthermore, the number of daily steps could be confirmed and used as a motivation for self-practice. In the future, longitudinal studies are needed to determine whether the ambulatory activity is maintained after discharge.

Conclusions
We found that cognitive function at admission was significantly associated with the daily step count at discharge in patients with subacute stroke who could walk independently. Therefore, patients with stroke with MCI may be at risk for poor ambulatory activity.