Introduction

Stroke is the second leading cause of global death and disability-adjusted life years (Roth et al., 2020). The number of deaths from stroke was 2.19 million in 2019 in China, and the disability-adjusted life years because of stroke amounted to about 45,949,100 person-years, accounting for 32.13% of the global disease burden and bringing a heavy social and economic burden (GBD 2019 Diseases and Injuries Collaborators, 2020). Guidelines recommend that all patients with stroke receive early rehabilitation interventions that involve physical therapy (PT) and occupational therapy (OT) to improve motor function and activities of daily living (ADLs) and promote recovery (Powers et al., 2018). However, few patients receive rehabilitation in the first week after a stroke (Asakawa et al., 2017). Rehabilitation professionals are restricted in terms of both numbers and specialization (Chimatiro & Rhoda, 2019). The still-unmet rehabilitation needs in many countries is a challenge addressed in the World Health Organization Rehabilitation 2030 agenda (Bernhardt et al., 2020). Moreover, the value of functional change in hospitals has been significantly related to functional recovery at 3 months after acute stroke (C. Y. Wang et al., 2021). Thus, acute-phase early rehabilitation requires greater attention.

One way to improve patient access to rehabilitation and address the shortage of rehabilitation therapists is to use “semisupervised practice,” which means engaging in rehabilitation practice without the supervision of therapists (Dorsch et al., 2019) and with the assistance of other healthcare professionals. Nurses regularly provide 24/7 care, have successfully narrowed the gap between the demand for and supply of care providers, and have expanded access to care for those who are disadvantaged (Chavez et al., 2018).

However, the current guidelines regarding including nurses in rehabilitation interventions during the acute phase are unclear, as current stroke rehabilitation guidelines reflect the healthcare situations in high-income countries, making the evidence translation challenging to apply to other settings (Platz, 2019). The responsibility challenges of nurses hinder the recovery of patients with stroke (Ramazanu et al., 2020). Thus, research exploring the potential roles and practices of nurses in acute rehabilitation in countries such as China is needed.

There is a lack of consensus regarding the role of nurses in stroke rehabilitation (Zerna et al., 2020). Researchers have previously summarized these roles to include conserving, interpretive, consoling, and integrative (Clarke, 2014). The integrative function requires nurses to target mobility and ADLs, integrating task-oriented training into daily activities to increase rehabilitation exercise training (Bjartmarz et al., 2017). Inpatient PT to increase muscle strength and deal with abnormal posture and active exercise time has been used only recently (Chimatiro & Rhoda, 2019). Although available to a small number of patients only, OT therapeutic activities and exercise to help patients improve self-care have been the most commonly used interventions (Shi & Howe, 2016). Nurses may learn PT or OT to play an integrated role.

The small amount of research on nurses' involvement in poststroke rehabilitation has generated inconsistent results (Meng et al., 2020). A nurse-led and caregiver-delivered stroke rehabilitation program was successfully implemented in rural China (Zhou et al., 2019). Another mobility program found that the number of nurse-led mobility sessions increased for a while after the intervention (Anton & Richard, 2018). Nevertheless, nurses redefined and expanded their role in rehabilitation care, contributing to an increased prevalence of rehabilitation exercises and training (Bjartmarz et al., 2017). Bridging this gap during the acute phase is recommended.

The Association of Rehabilitation Nurses has argued that nurses must integrate rehabilitation evidence at a professional level using the competency model (Vaughn et al., 2016). Moreover, this integration should use checklists or protocols in stroke care (Theofanidis & Gibbon, 2016) and use a theoretical framework to guide intervention program development.

Stroke-related physical dysfunction requires improving self-care abilities in rehabilitation therapy and daily activities. Orem's nursing system theory defines three systems of care, including fully compensatory, partially compensatory, and supportive education systems based on the degree of self-care deficits (Orem et al., 2001). In the partial compensatory system, patients with stroke should be recipients of rehabilitation services that mobilize their subjective abilities to gradually achieve self-care efficacy. When the patient can do some self-care activities, the nurse plays a role in the supportive education system. The role of nursing may be enhanced by introducing the Orem nursing system theory in the rehabilitation program for patients with stroke, selecting the appropriate nursing system for the patient, and stratifying patients' training levels based on the modified Barthel Index (MBI) score.

Program interventions promote functional recovery from stroke by accelerating the ability of patients to complete ADLs with PT and OT, preventing immobility-related complications such as joint stiffness and muscle atrophy, whereas early mobilization (EM) improves self-care abilities and reduces the length of stay in hospital or bed rest (Li et al., 2018). Developing an informative rehabilitation nursing program that contains these elements and arranges corresponding training for nurses is necessary.

This study aimed to examine the feasibility and effectiveness of a nurse-led motor function rehabilitation program. We hypothesized that the program intervention would improve patients' motor function and be feasible to integrate into practice.

Methods Design

This study included nursing program training and a 1:1 ratio parallel randomized controlled pilot trial. Reporting followed Consolidated Standards of Reporting Trials statement principles and the recommended checklist for randomized pilot and feasibility studies in the Consolidated Standards of Reporting Trials extension (Eldridge et al., 2016).

Participants

The inclusion criteria included (a) aged 18 years or older, (b) diagnosed with ischemic stroke and met the diagnostic criteria of the World Health Organization, (c) had initial stroke within 7 days with limb dysfunction, (d) National Institutes of Health Stroke Scale (NIHSS) scale consciousness scores of 0 or 1, and (e) provided informed consent form.

The exclusion criteria included (a) incomplete baseline clinical data; (b) experienced limb dysfunction for other reasons; (c) accompanied by deafness, aphasia, or severe cognitive impairment that precluded cooperating with the training program; and (d) having severe cardiopulmonary dysfunction or other diseases that required absolute bed rest. The flowchart describing the participant selection process is presented in Figure 1.

Figure 1.:

Flowchart of Study Participants

Nurses were screened for intervention implementation, with the following inclusion criteria: (a) > 5 years of neurology nursing experience, (b) experience as a nurse team leader, and (c) having provided consent to participate.

Program Development and Validation

We developed the motor function rehabilitation program using literature review and evidence summary methods and also integrated expert opinions from the multidisciplinary team. The MBI score was identified as the ideal criteria for grading and conducting function training, while the nurses determined individualized training items centered on patients.

This program was classified using MBI score into five levels, ranging from complete dependence (0–20) to basic self-care (80–100). The selection of nursing measures for patients was based on the nursing system theory of self-care (Orem et al., 2001). A score of 0–20 points indicated the use of a complete compensatory system, 21–60 points indicated a partial compensatory system, and over 60 points indicated a supportive-educative system. We referred to the competency model of professional rehabilitation nursing, identifying nurse-led interventions as the primary action to promote health and successful living, reflecting leadership, and playing a rehabilitative role (Vaughn et al., 2016). The program used supportive techniques, evidence-based assessment, and planning; established health-promoting goals; performed rehabilitation care programs; evaluated patient abilities; incorporated ethical and safe personalized care into the patient's rehabilitation goals and needs; participated in quality management activities and meetings; and promoted interprofessional team care. Nurses reinforced their leading role through the care process (assessment, diagnosis, planning, implementation, and evaluation) to ensure the recovery of patients with acute stroke.

Next, a structured consensus-based Delphi survey related to the program interventions was conducted. After the second consultation round, the questionnaire response rate was 100%, and the coefficient of expert authority was .90. The mean value of assignment importance was > 4.13, and Kendall's concordance coefficient was 0.226 (p < .001). Expert consensus was reached for all of the items. The methods used and the final edition of the five-grade 47-item rehabilitation nursing program have been published in Chinese (Jin et al., 2020).

Program Component

The interventions were variations of already established PT or OT. PT included limb and trunk movement, whereas OT included repetitive task-oriented training and ADL training to help patients improve their self-care ability. Many guidelines recommend EM such as sitting out of bed, transferring, and standing (Li et al., 2018). The program training items adopted in this study mainly involve the movement of the trunk, upper, and lower limbs as well as EM, task-oriented training, and ADL training (see Jin et al., 2020; J. Wang et al., 2020, for more details).

Guidelines recommend that training be of low dose and high frequency (Powers et al., 2018). Furthermore, as the mean length of stay in the acute phase of ischemic stroke in China is 10 (7.0–13.0) days (Y. J. Wang et al., 2020), the length of our intervention was set at 7 days. The intervention was conducted twice daily for 30 minutes each, as most training interventions range from 45 to 90 minutes per day (Stinear et al., 2020).

Implementation Procedure Preimplementation phase

Three months of program training and preparation preceded clinical application. The preimplementation phase consisted of the following four steps: (a) program initiation: identified program aims and target behaviors—all nurse participants were trained on case assessment, enrollment, study design, and data collection; (b) program mobilization: agreement on project and workshop plan as well as confirmation of site location and participants; (c) program implementation: training course (related to stroke, rehabilitation, and health education), provision of materials (presentation documents, text, pictures, animated videos, and other formatted documents), workshops (feedback or discussion), clinical practice, and examination; and (d) program closure: nurses were evaluated by specialized rehabilitation therapists and qualified to perform the relevant tasks, with no new problems arising in the clinical setting that would affect the program implementation.

Implementation phase

1. Sample size and randomization: We set the sample size for each group at 40 (1:1), as Hertzog (2008) recommended 30–40 samples per group for intervention efficacy pilot studies. A sample size of 100 cases was set, which allowed for a dropout rate up to 20%. The trained and qualified nurses enrolled participants and implemented the intervention (three in each ward). A researcher who did not participate in either the intervention or outcomes assessment generated a random number table and assigned participants to interventions. A minimum of 11 nurses were needed to conduct the intervention in this study.

2. Rehabilitation treatment: Patients received either a nurse-led motor function rehabilitation program (experimental group) or usual care (control group). The experimental group received training that included ADL training and EM such as bed and chair transfer, sitting balance, walking, and sit-to-stand. The intervention was performed within 7 days of stroke onset, with each session lasting 30 minutes for 7 consecutive days. Six intervention nurses were supervised by their department head nurse and randomly checked once weekly. Three head nurses attended the monthly meeting to review data quality and performance reports and discussed processes that need improvement.

The control group received usual rehabilitation care, including good limb placement, regular turning-over, and EM education following the hierarchical care system included in the Chinese National Health Council standards.

Data Collection

Outcome measures of effectiveness, feasibility, and acceptability were collected between January and October 2018.

Effectiveness indicators

We used the Motor Assessment Scale (MAS; scores of 0–48, interrater correlation of .95, and a test–retest correlation of .98) to evaluate patients' functional activity ability (Carr et al., 1985; Malouin et al., 1994), the MBI (scores of 0–100, reliability of .866–.997, criterion-related validity with Barthel Index of .816–1.000) to assess ADL (Min et al., 2008; Yang et al., 2022), and the NIHSS (scores of 0–42, reliability of .92–.96; Dewey et al., 1999) to determine the severity of poststroke neurologic deficit. These scales were assessed at baseline and 7 days after the intervention. For the MBI and MAS, higher scores on the scales indicate better functioning, whereas for the NIHSS, lower scales indicate better functioning. In this study, the intervention was considered effective when a significant difference in the value of outcome change between groups was found and the experimental group improved significantly better than the control group.

Feasibility indicators

First, the feasibility of the intervention was assessed using the percentage of patients who completed the study (a retention rate of ≥ 80% was required). Adverse events included stroke progression, cardiovascular complications, fall-induced injury, venous thromboembolism, pressure ulcers, pneumonia, urinary tract infections, and complications caused by improper exercise. These were monitored and prevented in the program implementation process using an incidence rate below or equal to 5% to indicate feasibility (AVERT Trial Collaboration Group, 2015). Patients with severe complications who were unstable stopped the intervention and were reassigned to the control group based on ethical principles. The fidelity of the intervention was described and ensured by the rehabilitation nurses who recorded the training items, time, duration, and number of item repetitions in the rehabilitation record form.

Acceptability indicators

A 10-item questionnaire scored using a 4-point Likert scale was developed in this study based on a review of the literature and an expert panel meeting, which investigated program acceptance among nurses and patients. Good program acceptance was determined by over 90% of participants agreeing with each questionnaire item.

Ethical Considerations

This study was approved by the human research ethics committee of the Second Affiliated Hospital of Zhejiang University School of Medicine (No. 2018-112) and registered in the ClinicalTrials.gov Protocol Registration and Results System. The study adhered to the Declaration of Helsinki, and all participants provided written informed consent.

Statistical Analysis

The normality of the data was tested first. Continuous variables were described using either means and standard deviations (SDs) or the median and interquartile range for nonnormal distributions. Categorical variables were described as frequency and proportions. Intragroup comparisons used the Wilcoxon signed-rank test to analyze whether the function had improved significantly after the intervention. Between-group comparisons used two independent samples t tests or the Mann–Whitney U test for continuous variables and chi-square tests for categorical variables. A p value of .05 (two-sided) was considered to be statistically significant. When the experimental group improved (changes measured by subtracting the baseline scores from the poststudy scores) significantly more than control group, the size of the difference was explored by calculating the effect sizes (r) using the formula Z/n, with Z value from the Mann–Whitney U test, n from the sample size, 0.5 for large effects, 0.3 for medium effects, and 0.1 for small effects (Fritz et al., 2012). Data were analyzed on IBM SPSS Statistics 22.0 (IBM Inc., Armonk, NY, USA) software.

Results Baseline Characteristics of Participants

From January to October 2018, 104 patients met the inclusion criteria, 101 completed the baseline assessments, 88 (49 men; mean age = 62.8 years [SD = 13.2]) completed both assessments, and data from 43 and 45 in the experimental and control groups, respectively, were available for analysis (Figure 1). We compared the baseline demographic and clinical data of the 88 participants. The median enrollment time was 3–4 days after stroke onset, and the median muscle strength values in the affected upper and lower limbs were 2 and 3, respectively. The median MBI score was 36 (27, 49) for the experimental group and 39 (25, 51) for the control group, indicating moderate-to-severe dependence in ADLs. The median MAS score was 16 for both groups, and the median NIHSS scores were 6 (4, 8) for the experimental group and 7 (3, 10) for the control group, indicating moderate-to-severe motor dysfunction and neurological impairment. No statistically significant between-group differences were observed at baseline (Table 1).

Table 1. - Comparison of Baseline Demographic and Clinical Data Variable Experimental Group (n = 43) Control Group (n = 45) t/Z/χ2 p n % n % Age (years; M and SD) 64.35 10.59 61.36 15.27 t = 1.073 .287 Male 23 46.9 26 53.1 0.164 .686 Onset to enrollment (days; M and SD) 3.81 1.61 4.07 1.81 t = −0.690 .492 Affected limb muscle strength (median and IQR)  Upper 2.0 1.0–4.0 2.0 0.5–4.0 Z = −0.290 .772  Lower 3.0 2.0–4.0 3.0 2.0–4.0 Z = −0.659 .510 Educational background 5.747 .219  Illiterate 10 23.3 7 15.6  Primary school 12 27.9 14 31.1  Junior high school 8 18.6 17 37.8  High school and above 13 30.2 7 15.5 Medical payment 2.228 .328  Medical insurance for urban workers 8 18.6 5 11.2  Medical insurance for urban residents 13 30.2 20 44.4  Self-paying 22 51.2 20 44.4 Risk factorsa  Smoking 17 39.5 19 42.2 0.066 .798  Family history of CVA or TIA 8 18.6 8 17.7 0.010 .920  Hypertension 24 55.8 32 71.1 2.224 .136  Diabetes mellitus 11 25.6 14 31.1 0.331 .565  Atrial fibrillation or coronary heart disease 4 9.2 3 6.7 0.004 .950 Stroke severity (NIHSS; median and IQR) 6.0 4.0–8.0 7.0 3.0–10.0 Z = −0.482 .630 MBI score (median and IQR) 36.0 27.0–49.0 39.0 25.0–51.0 Z = −0.008 .993 MAS score (median and IQR) 16.0 7.0–29.0 16.0 8.0–26.5 Z = −0.067 .947

Note. IQR = interquartile range; CVA = cerebrovascular accident; TIA = transient ischemic attack; NIHSS = National Institutes of Health Stroke Scale; MBI = modified Barthel Index; MAS = Motor Assessment Scale.

aMultiple choice.

Effectiveness

Both groups improved significantly between preintervention and postintervention in terms of MBI, MAS, and NIHSS (p < .001; Table 2). The Mann–Whitney U test showed that the experimental group improved significantly more than the control group in terms of MBI (p = .002) and MAS (p = .001). However, the difference between the two groups was not statistically significant in terms of NIHSS (p = .673). The differences were of medium effect size for MBI (r = .32) and MAS (r = .34; Table 3).

Table 2. - Intragroup Comparison of Outcomes Pretest vs. Posttest Outcome/Assessment Experimental Group (n = 43)
Median and IQR Control Group (n = 45)
Median and IQR Pre Post p Pre Post p Modified Barthel Index 36.0 27.0–49.0 62.0 42.0–87.0 < .001 39.0 25.0–51.0 54.0 32.5–71.0 < .001 Motor Assessment Scale 16.0 7.0–29.0 30.0 15.0–42.0 < .001 16.0 8.0–26.5 24.0 10.0–33.5 < .001 NIHSS 6.0 4.0–8.0 3.0 1.0–6.0 < .001 7.0 3.0–10.0 5.0 2.5–8.0 < .001

Note. Statistical test: Wilcoxon signed-rank test. IQR = interquartile range; NIHSS = National Institutes of Health Stroke Scale.

Table 3. - Between-Group Comparison of Outcome Score Changes (Posttest–Pretest) Outcome Median and IQR Z p Effect Size
(n = 88) Experimental Group
(n = 43) Control Group
(n = 45) Modified Barthel Index 20.0 12.0–32.0 11.0 6.0–19.5 −3.046* .002 0.32 Motor Assessment Scale 9.0 5.0–13.0 5.0 2.0–7.0 −3.264* .001 0.34 NIHSS 1.0 −2.0 to 2.0 0.0 −0.5 to 1.0 −0.422 .673 0.04

Note. Statistical test: Mann–Whitney U test. IQR = interquartile range; NIHSS = National Institutes of Health Stroke Scale.

*p < .05.

Feasibility

As shown in Figure 1, 43 (86%) in the experimental group completed the 7-day intervention and 45 (88%) in the control group completed the full assessment, showing a high and comparable retention rate between the two groups. The main reason for dropping out of the study was early discharge from the hospital. Two participants with stroke progressions in the experimental group changed to usual care during the study period. None of the patients experienced severe adverse events during the study period. There was no lack of information in the rehabilitation record form, and no training was reported to be inconsistent with the intervention program during the weekly random checks conducted by head nurses. Thus, fidelity of nurses in the delivery of intervention was well ensured.

Acceptability

Eleven nurses were recruited and trained (six charge nurses, three head nurses, and two nurse researchers in three neurology wards). The questionnaire response rate was 100% for the nurses. Of the 11 nurses, 36.4% and 63.6% agreed or totally agreed, respectively, that the program training was well planned and organized (Item 1) and stroke rehabilitation strengthened their role and function (Item 4). Nearly three quarters of the nurses (72.7%) totally agreed that the program was educational (Item 2) and that they had mastered the requisite stroke rehabilitation skills (Item 5). Furthermore, slightly over half (54.5%) of the nurse participants totally agreed that the program could be effectively integrated into their daily practice (Item 3; Table 4).

Table 4. - Acceptability of Program Implementation Item Totally Agree Agree Disagree n % n % n % For nurses 1. This training was well planned and organized. 7 63.6 4 36.4 0 0.0 2. This training can improve stroke rehabilitation knowledge and skills. 8 72.7 3 27.3 0 0.0 3. The program can be well integrated into nursing practice. 6 54.5 5 45.5 0 0.0 4. The program strengthened the role and function of nursing in stroke rehabilitation. 7 63.6 4 36.4 0 0.0 5. After clinical application, you have mastered the operation of and can guide for stroke rehabilitation skills. 8 72.7 3 27.3 0 0.0 For patients 6. The training service, procedures, and environment were good. 29 67.4 14 32.6 0 0.0 7. You felt involved and entertained during the training. 18 41.9 22 51.2 3 7.0 8. You and your caregiver can easily integrate rehabilitation skills into your activities of daily living. 16 37.2 25 58.1 2 4.7 9. The training was necessary to improve stroke rehabilitation knowledge and skills. 32 74.4 10 23.3 1 2.3 10. Nurses can play a vital role and function in stroke rehabilitation. 28 65.1 15 34.9 0 0.0

Note. As no “totally disagree” responses were received for any of the 10 items, this column is not displayed here.

The questionnaire response rate was 86% for the participants (43 in the experimental group). All of those who had received the intervention evaluated the training service, procedures, and environment as good (Item 6) and agreed that nurses can play a vital role and function in stroke rehabilitation (Item 10). Seven percent of the participants self-reported as having a poor experience with the training (Item 7). Nearly all (95.3%) of the patients and caregivers agreed they could easily integrate the rehabilitation skills into ADLs (Item 8), and most of the patients perceived the training as necessary (Item 9).

Discussion

This study was the first time that this nurse-led motor function rehabilitation program has been scientifically evaluated. MAS and MBI scores showed greater improvement in the experimental group. The program was found to have a high retention rate and good fidelity. Nurses and patients considered the program acceptable and expressed that it would fit well into their daily practice. In general, the rehabilitation nursing program may be considered to be feasible and preliminarily effective for patients with acute stroke.

The nurse-led motor function rehabilitation program was found to be preliminarily effective. First, the changes in MBI scores in the experimental group were more significant than those in the control group, which may be explained by the adjustment of rehabilitation nursing care based on MBI level. ADLs wer