This paper is in the following e-collection/theme issue:

mHealth for Data Collection and Research (981) mHealth for Symptom and Disease Monitoring, Chronic Disease Management (1428) mHealth in a Clinical Setting (624) Cognitive and Neurorehabilitation (267) Telerehabilitation (162) Stroke (166)

Published on in Vol 8, No 5 (2020): May

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/17219, first published .
A Mobile Health App for the Collection of Functional Outcomes After Inpatient Stroke Rehabilitation: Pilot Randomized Controlled Trial

A Mobile Health App for the Collection of Functional Outcomes After Inpatient Stroke Rehabilitation: Pilot Randomized Controlled Trial

A Mobile Health App for the Collection of Functional Outcomes After Inpatient Stroke Rehabilitation: Pilot Randomized Controlled Trial

Authors of this article:

Li Li1 Author Orcid Image ;   Jia Huang1, 2 Author Orcid Image ;   Jingsong Wu1 Author Orcid Image ;   Cai Jiang1 Author Orcid Image ;   Shanjia Chen1 Author Orcid Image ;   Guanli Xie1 Author Orcid Image ;   Jinxin Ren1 Author Orcid Image ;   Jing Tao1, 2, 3 Author Orcid Image ;   Chetwyn C H Chan4 Author Orcid Image ;   Lidian Chen1, 2, 3 Author Orcid Image ;   Alex W K Wong5, 6, 7 Author Orcid Image
Article Authors Cited by (14) Tweetations (19) Metrics

    Original Paper

    1College of Rehabilitation Medicine, Fujian University of Traditional Chinese Medicine, Fuzhou, China

    2Key Laboratory of Orthopedics & Traumatology of Traditional Chinese Medicine and Rehabilitation, Fujian University of Traditional Chinese Medicine, Ministry of Education, Fuzhou, China

    3Traditional Chinese Medicine Rehabilitation Research Center of State Administration of Traditional Chinese Medicine, Fuzhou, China

    4Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, HongKong, Hong Kong

    5Program in Occupational Therapy, Washington University School of Medicine, St. Louis, MO, United States

    6Department of Neurology, Washington University School of Medicine, St. Louis, MO, United States

    7Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, United States

    *these authors contributed equally

    Corresponding Author:

    Alex W K Wong, PhD, DPhil

    Program in Occupational Therapy

    Washington University School of Medicine

    600 South Taylor Avenue, MSC 8505-94-01

    St. Louis, MO

    United States

    Phone: 1 314 286 0278

    Email: wongal@wustl.edu


    Background: Monitoring the functional status of poststroke patients after they transition home is significant for rehabilitation. Mobile health (mHealth) technologies may provide an opportunity to reach and follow patients post discharge. However, the feasibility and validity of functional assessments administered by mHealth technologies are unknown.

    Objective: This study aimed to evaluate the feasibility, validity, and reliability of functional assessments administered through the videoconference function of a mobile phone–based app compared with administration through the telephone function in poststroke patients after rehabilitation hospitalization.

    Methods: A randomized controlled trial was conducted in a rehabilitation hospital in Southeast China. Participants were randomly assigned to either a videoconference follow-up (n=60) or a telephone follow-up (n=60) group. We measured the functional status of participants in each group at 2-week and 3-month follow-up periods. Half the participants in each group were followed by face-to-face home visit assessments as the gold standard. Validity was assessed by comparing any score differences between videoconference follow-up and home visit assessments, as well as telephone follow-up and home visit assessments. Reliability was assessed by computing agreements between videoconference follow-up and home visit assessments, as well as telephone follow-up and home visit assessments. Feasibility was evaluated by the levels of completion, satisfaction, comfort, and confidence in the 2 groups.

    Results: Scores obtained from the videoconference follow-up were similar to those of the home visit assessment. However, most scores collected from telephone administration were higher than those of the home visit assessment. The agreement between videoconference follow-up and home visit assessments was higher than that between telephone follow-up and home visit assessments at all follow-up periods. In the telephone follow-up group, completion rates were 95% and 82% at 2-week and 3-month follow-up points, respectively. In the videoconference follow-up group, completion rates were 95% and 80% at 2-week and 3-month follow-up points, respectively. There were no differences in the completion rates between the 2 groups at all follow-up periods (X21=1.6, P=.21 for 2-week follow-up; X21=1.9, P=.17 for 3-month follow-up). Patients in the videoconference follow-up group perceived higher confidence than those in the telephone follow-up group at both 2-week and 3-month follow-up periods (X23=6.7, P=.04 for 2-week follow-up; X23=8.0, P=.04 for 3-month follow-up). The videoconference follow-up group demonstrated higher satisfaction than the telephone follow-up group at 3-month follow-up (X23=13.9; P=.03).

    Conclusions: The videoconference follow-up assessment of functional status demonstrates higher validity and reliability, as well as higher confidence and satisfaction perceived by patients, than the telephone assessment. The videoconference assessment provides an efficient means of assessing functional outcomes of patients after hospital discharge. This method provides a novel solution for clinical trials requiring longitudinal assessments.

    Trial Registration: chictr.org.cn: ChiCTR1900027626; http://www.chictr.org.cn/edit.aspx?pid=44831&htm=4.

    JMIR Mhealth Uhealth 2020;8(5):e17219

    doi:10.2196/17219

    Keywords

    telemedicine; cell phone; stroke; rehabilitation; activities of daily living; outcome and process assessment; health care 


    Stroke is one of the leading causes of death and disability worldwide [1]. As in other developing countries, the incidence and prevalence of stroke in China are gradually increasing. Each year, China has 2.5 million new stroke cases and more than 11 million stroke survivors; stroke has become the leading cause of death in China [2]. Stroke can have a long-term impact on an individual’s physical, mental, and social function, as well as on survivors’ caregivers and families [3-6]. The majority of patients receive inpatient stroke rehabilitation to regain function for only the first few weeks after stroke, but functional recovery often occurs 3 months or even longer following a stroke [7]. Additionally, about 30% of poststroke individuals receive outpatient rehabilitation [8]. Even if available, access to rehabilitation for patients in China is limited because of transportation, geographical barriers, and monetary factors [9]. As most patients have recovery potential but do not receive recommended rehabilitation, it is important to develop new strategies to continuously monitor functional recovery and other health outcomes of patients following discharge to better understand the long-term consequences for patients poststroke [10].

    Poststroke home-based therapies seem to be a viable option for the delivery of stroke care. Follow-up assessments and interventions not only provide a means of monitoring the functional status of patients after transitioning to home and community [11], but also provide instructions to prevent readmission, which is especially important for those who receive a longer stay in inpatient rehabilitation [12]. Moreover, follow-up assessments enable clinicians to adjust the treatment plan for home-based therapies [13,14]. One common method for follow-up assessment is a face-to-face, at-home assessment in which the home health therapist visits the patient at home to perform an evaluation. However, this method demands intensive resources, including the time of trained personnel and financial expenditures [15]. Recent studies have tested alternative methods of follow-up data collection for patients following a stroke [16-18].

    Telephone administration is a common alternative. This method allows participants to be recruited from diverse geographical areas, is typically less expensive than the face-to-face home assessment, and has a quick turnaround time [19]. Prior studies have found that telephone administration of outcome measures demonstrates equivalent reliability to face-to-face assessment, supporting telephone interview as a feasible solution [20,21]. However, there are some shortcomings to telephone administration. First, many functional measures require a trained therapist to observe and provide ratings of how the patient performs specific daily tasks. Inability to perform observations via telephone administration may be a hindrance to accurately evaluating task performance. Furthermore, telephone administration often assesses survey-based questions, which require patients to have higher education, health literacy, and communication abilities to understand the verbal instructions [22]. An earlier study found a large amount of missing data from assessments administered through the telephone interview method for stroke patients and caregivers, limiting the use of this method in clinical trials requiring longitudinal assessments [23].

    With advances in computing power and mobile connectivity, many mobile health (mHealth) technologies, such as mobile devices, sensors, apps, and social media, are becoming available to obtain data pertinent to wellness and disease diagnosis, prevention, and management [24]. WeChat (Chinese version: Weixin), developed in 2011 by Tencent, has become the most common social software app in China [25]. Similar to other social media apps, such as Facebook, Twitter, and WhatsApp, WeChat is a free platform that provides seamless opportunities for communication and other mobile apps. People can communicate with one another through the free voice call or video call feature, as well as instantly share information [26]. According to the Statista Research Department [27], WeChat had over 1.15 billion monthly active users from a wide range of age groups. Harnessing the use of mHealth technologies to improve health and wellness is not uncommon in modern health care [28]. A recent mHealth intervention study utilizing the WeChat app for weight loss behaviors in a group of male workers found promising results: participants who spent more time using the health education program embedded in the WeChat app for engaging in healthy behaviors demonstrated more weight loss [29]. Another mHealth study used the WeChat app to educate parents of pediatric patients undergoing surgery and found that this mobile app–assisted intervention was effective in enhancing parents’ knowledge of perioperative procedures [30]. Another study used the WeChat app in a group of discharged patients with head and neck tumors for 6 months and demonstrated the app to be a cost-effective method of follow-up assessment [31].

    Although the utility of the WeChat app has been demonstrated in other populations, little is known about whether the WeChat app could be feasible to assess the postdischarge functional status of patients following a stroke, because most poststroke patients experience cognitive and communication difficulties that may make it difficult to operate the app and understand its instructions. To address this question, we conducted a pilot randomized controlled trial in a group of discharged stroke patients by randomly assigning them into 2 different modes of administration during 2-week and 3-month follow-up periods: WeChat video conference or WeChat telephone administration. This study had 2 specific aims. The first aim was to compare the validity and reliability of functional assessment between these 2 modes of administration in stroke patients. We hypothesized that videoconference administration would demonstrate higher validity and reliability than telephone administration, because examiners using videoconference administration can observe how the respondent performs specific tasks to provide appropriate ratings, whereas examiners in the telephone administration group demand more subjective appraisals of task performance based on the respondent’s verbal descriptions. Our second aim was to examine the feasibility of the functional assessment administered via the videoconference function compared with the telephone call function in stroke patients after rehabilitation hospitalization. We hypothesized that both modes of administration would demonstrate high levels of completion, comfort, satisfaction, and confidence.


    Study Design

    This study was a parallel, 2-group, and pragmatic randomized controlled trial of an mHealth app of functional outcome data collection after inpatient stroke rehabilitation. The trial was registered with the Chinese Clinical Trial Registry: ChiCTR1900027626. A total of 120 eligible stroke patients from the affiliated rehabilitation hospital of Fujian University of Traditional Chinese Medicine were recruited for this study. Participants were approached by a research assistant, who provided the study information. After participants provided informed consent and were screened for eligibility, participants were randomized into 1 of the 2 WeChat app administration groups: videoconference follow-up or telephone follow-up, with a ratio of 1:1. Eligible participants were randomized using a random number table generated by a study coordinator who was not involved in the recruitment and assessment of participants for the study.

    Ethics Approval

    This trial was implemented in compliance with the declaration of Helsinki and approved by the Ethics Board of the affiliated rehabilitation hospital of Fujian University of Traditional Chinese Medicine (number: 2016KY-032-01). All participants provided informed, written consent before participation. Participants received an honorarium to acknowledge their research contribution.

    Participants

    Patients were eligible to participate in the study if they met the following criteria: (1) aged at least 18 years, (2) diagnosis of first stroke, (3) normal speech function according to the Mandarin Language Screening Test with cutoff scores of >13 for those with primary school education or >14 for those with junior high school or higher education, (4) normal cognitive function according to the Montreal Cognitive Assessment with a cutoff score of >26, and (5) home discharge. Participants were excluded if they (1) did not own a mobile phone, (2) were unwilling to install and use the WeChat software on their mobile phone, (3) had emotional dysfunction according to the Beck Depression Inventory with a cutoff score of >13, or (4) had other medical illnesses limiting study participation. As we included only participants who served as their own informant rather than including participants on a nonselected basis, it is likely that individuals who were too cognitively impaired or were unable to understand the study materials were excluded.

    Recruitment and Screening

    Participants were recruited from the inpatient rehabilitation hospital. Recruitment was initiated while the stroke patient was still in the hospital. The research assistant screened the medical records of all patients undergoing inpatient rehabilitation for a stroke. Once potential participants were identified, the research assistant approached the individual and provided information about the study. Participants provided informed consent once they agreed to participate. The research assistant reviewed the inclusion and exclusion criteria and completed the screening tests to confirm eligibility. Participants were then enrolled, and randomization occurred only after recruitment by a study coordinator.

    Data Collection Procedures

    All research assistants received training in assessing the eligibility of potential participants, obtaining informed consent from participants, the study protocol, and obtaining outcome measures for both groups. They also received training in how to coach and assist participants using the WeChat app, including the videoconference and telephone call functions. After randomization, all participants received the baseline assessment at the week of discharge, followed by the completion of 2 mHealth app follow-up sessions (either videoconference or telephone), and half of the participants from each group received 2 home visits. The first follow-up session occurred 2 weeks after home discharge. Within 1 week of the first follow-up session, half of the study participants were selected to conduct the first home visit based on stratified sampling in each group. The stratified sampling criteria were grounded on participants’ functional abilities. As home visits are costly, this study was limited by randomly selecting half of the study participants for the home visit assessment. The second follow-up session occurred 3 months after home discharge. Within 1 week of the second follow-up session, we completed the second home visit in this subgroup of participants. The time interval between videoconference or telephone follow-up and home visit of 1 week was considered long enough to ensure that the previous responses were forgotten and short enough to ensure that the patient’s clinical condition would not substantially change. All assessments were conducted by our trained research assistants with training in physical therapy, occupational therapy, or rehabilitation medicine. To reduce assessor bias in the trial, research assistants who completed the videoconference or telephone follow-up sessions also conducted home visits with patients in the same group. We treated the face-to-face, home visit assessment as the gold standard in this study.

    Participants in the videoconference follow-up group received training on the usage of the videoconference function of the WeChat app before discharge. During the follow-up sessions, research assistants asked participants to complete individual functional tasks and rate their actual performance through the videoconference, with the exception of bladder and bowel management tasks, which were assessed by the participant’s verbal descriptions. Participants also described difficulties pertaining to their individual task performance. During the home visit, research assistants completed the face-to-face observations by rating participants as they completed the same functional tasks. We used the same scoring criteria to evaluate the performance of our study participants in both videoconference and face-to-face, home visit assessments. Participants in the telephone follow-up group received training on the usage of the telephone function of the WeChat app before discharge. During the follow-up sessions, research assistants made telephone calls, asked participants how they performed in each functional item, and appraised their performance based on the participant’s verbal descriptions. During the home visit, research assistants completed the same protocol as in the videoconference follow-up group. We used the same scoring criteria to evaluate the performance of our study participants in both telephone and home visit assessments.

    Outcome Measures

    As recommended in a systematic review of optimal outcome measures for stroke therapy trials [32], the primary outcome selected for this study was improvement in activities and participation rather than the reduction of impairments. Thus, we chose the functional status (ie, the performance of activities of daily living, ADLs) of stroke participants as our primary outcome variable. We used the Modified Barthel Index (MBI) to assess the performance of ADLs in the 2 groups [33,34]. The MBI can be administered using clinician-rated or patient-reported methods. It includes 10 items measuring grooming, bathing, feeding, toileting, stair climbing, dressing, bowel control, bladder control, mobility, and chair/bed transfer. Items have different response options, with anchored scores provided for different options. The total score ranges from 0 to 100. A higher score means that the participant has greater independence. Adequate validity and reliability were found for the Chinese version of the MBI used in this study [35].

    We also defined 2 outcome variables to examine the feasibility of using an mHealth app to measure functional status of stroke participants in this study. The first variable was the completion of the MBI among our study participants in both groups at both follow-up periods. The second variable was the acceptability among our study participants (ie, levels of satisfaction, comfort, and confidence) of using the videoconference or telephone functions to complete the functional assessment at follow-up periods. We developed 3 questions: (1) “Are you satisfied with this follow-up assessment?” (2) “Are you comfortable with this follow-up assessment?” and (3) “Are you confident using this follow-up assessment?” All items were rated on a 4-point scale ranging from “very satisfied/comfortable/confident” to “unsatisfied/uncomfortable/unconfident.”

    Statistical Analysis

    Data were analyzed using the Statistical Package for Social Sciences, version 20 (IBM, Chicago, IL, United States). Baseline characteristics between groups were compared using t tests or Mann-Whitney tests for continuous variables, and Pearson chi-square tests for categorical variables. We applied Wilcoxon signed-rank tests to compare MBI score differences between videoconference and telephone assessments, as well as between videoconference/telephone and home visit assessments for the validity evaluation. We set the P value to .05 for statistical significance. We used intraclass correlation coefficients (ICCs) to evaluate the agreement of all item scores between these assessments for the reliability evaluation. According to Landis and Koch [36], the ICC can theoretically vary between 0 and 1.0, where an ICC of 0 indicates no reliability, and an ICC of 1.0 indicates perfect reliability; ICCs above 0.80 indicate acceptable reliability. We used chi-square tests or Fisher exact tests to compare rates of completion, satisfaction, comfort, and confidence between the videoconference and telephone modes of administration.


    Baseline Characteristics

    Figure 1 illustrates the flow of participant enrollment. Among 519 potential stroke inpatients, 353 did not meet the inclusion criteria. A total of 21 patients refused to participate in the study because family members were uncertain about the use of mHealth for collecting data. Some patients reported that they could easily access medical services and did not require additional follow-up services. In total, 25 patients were discharged from the hospital before research assistants approached them. A total of 120 participants were successfully recruited and randomized to 1 of the 2 groups. Table 1 describes the demographic characteristics of study participants. Study participants were middle-aged (mean age 59.7 years, SD 12.1), 59.1% (71/120) of participants were women, and 93.3% (112/120) of participants were married. In total, 45.8% (55/120) of participants completed 9 or fewer years of formal education, and 28.3% (34/120) of participants were currently employed. A total of 45.0% (54/120) of participants had a history of cerebral infarction (ie, ischemic stroke). We found no significant differences between the 2 groups on gender, marital status, education, occupation, type of stroke, or duration of disease. We also found no significant differences between the 2 groups in any functional task measured by the MBI at the time of discharge. Eight participants in the videoconference follow-up group dropped out at 2 weeks: 3 participants did not answer the video calls and 5 participants refused to complete the assessment. Three participants in the telephone follow-up group dropped out at 2 weeks; they did not answer the telephone calls. At the 3-month follow-up point, we lost more participants: 3 participants in the videoconference follow-up group and 9 in the telephone follow-up group. Of the initial 60 participants in each group, 82% (49/60) of participants in the videoconference follow-up group and 80% (48/60) of participants in the telephone follow-up group completed the entire study protocol.

    ‎
    Figure 1. A flow diagram of patient enrollment.
    View this figure
    Table 1. Baseline characteristics of participants.
    VariablesAll participants (N=120)Telephone follow-up (n=60)Videoconference follow-up (n=60)Z test or Χ2 (df) testP value
    Age (years), mean (SD)59.7 (12.1)60.8 (11.6)58.3 (12.4)1.56.25
    Gender, n (%)0.14 (1).85

    		Male49 (40.8)24 (41.7)25（36.7）


    		Female71 (59.1)36（58.3）35（63.3）

    Marital status, n (%)0.53 (1).72

    		Married112 (93.3)57（95.0）55（91.7）


    		Other8 (6.7)3（5.0）5（8.3）

    Education (years), n (%)1.38 (3).71

    		≤625 (20.8)14（23.3）11（18.3）


    		7-941 (34.2)22（36.7）19（31.7）


    		9-1230 (25.0)14（23.3）16（26.7）


    		≥1224 (20.0)10（16.7）14（23.3）

    Occupation, n (%)1.57 (3).67

    		Employed34 (28.3)19（31.7）15（25.0）


    		Retired50 (41.7)26（43.3）24（40）


    		Unemployed10 (8.3)4（6.7）6（10）


    		Other26 (21.7)11（18.3）15（25）

    Type of stroke, n (%)2.15 (1).14

    		Infarction54 (45.0)23（38.3）31（51.7）


    		Hemorrhage66 (55.0)37（61.7）29（48.3）

    Duration of stroke (days), mean (SD)90.7 (13.8)87.6 (14.6)93.7 (12.9)0.53.24
    Discharge functional status (Modified Barthel Index), mean (SD)

    		Feeding7.27 (2.09)7.35 (1.92)7.12 (2.08)–0.59.55

    		Grooming3.06 (0.83)3.02 (0.68)3.15 (0.90)–1.64.10

    		Dressing5.32 (2.10)5.08 (2.09)5.62 (2.27)–1.35.18

    		Bathing2.02 (1.46)2.23 (1.09)1.85 (1.15)–2.02.06

    		Toilet use5.21 (2.24)5.13 (1.82)5.32 (2.14)–0.49.62

    		Bowels8.87 (1.64)8.65 (1.63)9.08 (1.33)–1.30.19

    		Bladder9.06 (1.17)9.00 (1.24)9.18 (1.11)–1.24.22

    		Transfer8.57 (2.89)8.40 (2.68)8.70 (3.03)–0.68.49

    		Mobility7.73 (2.89)8.15 (2.92)7.42 (3.31)–1.34.18

    		Stairs4.66 (2.17)4.48 (1.87)4.83 (2.27)–0.77.44

    Comparison of Videoconference and Telephone Follow-Up Assessments

    Table 2 shows the MBI scores for videoconference follow-up and telephone follow-up at 2-week and 3-month periods. We found no significant differences between the 2 groups in the majority of functional tasks measured by the MBI at 2 weeks and 3 months, except that significant differences were found in the bladder management task at 2 weeks, and the grooming and bathing tasks at 3 months. ICC values for all but the grooming task at 2 weeks and the grooming, toilet use, and mobility tasks at 3 months exceeded 0.8, indicating acceptable reliability between videoconference and telephone assessments at both follow-up periods.

    Table 2. Comparison of Modified Barthel Index scores evaluated by videoconference and telephone follow-up assessments at 2-week follow-up and 3-month follow-up.
    VariablesTelephone follow-up, mean (SD)Videoconference follow-up, mean (SD)Z testP valueIntraclass correlation coefficient
    2-week follow-up

    		Feeding8.64 (1.59)8.44 (1.72)–1.64.100.82

    		Grooming3.65 (0.69)3.45 (0.76)–1.79.080.74

    		Dressing6.74 (2.00)6.44 (2.03)–1.90.060.83

    		Bathing3.07 (1.14)2.72 (1.22)–1.38.150.86

    		Toilet use7.13 (2.13)6.28 (1.84)–1.72.070.81

    		Bowels8.63 (1.50)9.03 (0.99)–1.53.120.88

    		Bladder9.04 (1.66)9.45 (0.89)–2.06.040.83

    		Transfer9.50 (2.63)9.85 (2.78)–1.46.130.82

    		Mobility8.75 (2.90)8.25 (2.99)–0.79.420.85

    		Stairs6.67 (2.05)6.28 (1.83)–0.58.560.81
    3-month follow-up

    		Feeding8.93 (1.59)8.63 (1.38)–1.77.080.85

    		Grooming4.28 (0.80)3.93 (0.58)–2.85.040.65

    		Dressing7.33 (2.28)6.68 (1.81)–1.86.060.81

    		Bathing3.30 (0.99)2.80 (1.09)–1.93.050.82

    		Toilet use7.25 (1.94)6.78 (1.87)–1.25.210.76

    		Bowels9.23 (1.21)9.43 (0.91)–0.70.480.81

    		Bladder9.03 (1.59)9.50 (0.87)–1.29.200.83

    		Transfer10.38 (2.74)10.47 (2.70)–0.32.750.82

    		Mobility10.32 (3.13)9.45 (2.99)–1.49.140.66

    		Stairs7.05 (2.05)6.48 (2.33)–1.33.180.80

    Comparison of Two Mobile Health Follow-Up Assessments With Home Visit Assessments

    Selecting Candidates for Home Visit Assessments in Two Groups

    We adopted stratified sampling in each group to select subgroups of participants for home visit assessments. We used the discharge MBI scores of 52 participants in the videoconference follow-up group and 57 participants in the telephone follow-up group to classify their functional independence levels into 4 videoconference follow-up subgroups and 4 telephone follow-up subgroups, respectively. The number of participants in the videoconference follow-up and telephone follow-up subgroups is shown in Table 3. We selected half of the participants in each of the 8 subgroups for home visit assessments.

    Table 3. Distribution of discharge Modified Barthel Index scores in the videoconference follow-up and telephone follow-up subgroups.
    MBIa scoresVideo follow-up, n (%)Telephone follow-up, n (%)
    Complete dependence (MBI<40)3 (5.8)4 (7)
    Dependence (MBI 40-59)17 (32.7)16 (28)
    Mild dependence (MBI 60-99)30 (57.7)36 (63.1)
    Independence (MBI 100)2 (3.8)1 (1.9)

    aMBI: Modified Barthel Index.

    Comparison of Modified Barthel Index Scores Between Videoconference Follow-Up and Home Visit

    Table 4 shows the MBI scores for videoconference follow-up and face-to-face, home visit assessments at 2-week follow-up. MBI scores collected by videoconference were similar to those collected by the face-to-face, home visit assessment, except that a significant difference was found in the feeding task. We also found that ICC values for all 10 tasks were above 0.8, indicating acceptable reliability between the 2 assessments at 2-week follow-up. Table 4 also shows the MBI scores for videoconference follow-up and home visit assessments at 3-month follow-up. Similar comparison results were found between the 2 assessments at 3 months, except that a significant difference was found in the transfer task. ICC values for all but the transfer and stair-climbing tasks exceeded 0.8, indicating acceptable reliability between videoconference and home visit assessments at 3 months.

    Table 4. A comparison of MBI scores evaluated in videoconference and home visit assessments at 2-week follow-up (n=26) and 3-month follow-up (n=25).
    VariablesVideoconference, mean (SD)Home visit, mean (SD)Z testP valueIntraclass correlation coefficient
    2-week follow-up

    		Feeding8.04 (1.75)7.23 (1.95)−2.60<.0010.87

    		Grooming3.30 (0.88)3.50 (0.76)−1.89.060.90

    		Dressing6.35 (2.21)6.58 (2.21)−1.00.320.92

    		Bathing2.46 (1.27)2.54 (1.24)0.49.620.87

    		Toilet use6.23 (2.02)5.77 (2.08)−1.63.100.87

    		Bowels9.30 (0.97)9.23 (0.99)−0.58.560.86

    		Bladder9.46 (0.90)9.36 (0.96)−0.09.870.89

    		Transfer9.27 (2.84)9.15 (3.51)−0.72.470.84

    		Mobility8.96 (3.18)9.30 (2.99)1.0.320.87

    		Stairs6.42 (1.96)5.81 (1.83)−1.83.070.82
    3-month follow-up

    		Feeding8.54 (1.42)8.12 (1.80)–1.73.080.83

    		Grooming4.04 (0.53)4.12 (0.65)–1.00.320.88

    		Dressing6.77 (1.75)7.04 (1.56)–1.51.130.92

    		Bathing2.77 (1.10)3.00 (1.13)–1.35.180.82

    		Toilet use6.77 (1.95)7.00 (1.90)0.81.420.80

    		Bowels9.31 (0.97)9.38 (0.94)–0.58.560.89

    		Bladder9.46 (0.90)9.54 (0.86)–1.00.320.95

    		Transfer10.77 (2.80)9.50 (3.31)–2.41.020.75

    		Mobility9.81 (2.23)9.88 (2.76)0.14.890.85

    		Stairs6.88 (2.10)6.35 (1.65)–1.48.140.76

    Comparison of Modified Barthel Index Scores Between Telephone Follow-Up and Home Visit

    Table 5 shows the MBI scores for telephone follow-up and face-to-face, home visit assessments at 2-week follow-up. A comparison of these assessments found that almost all MBI scores collected by telephone administration were statistically higher than those collected by the home visit assessment, with the exception of bowel and bladder management tasks, indicating that the telephone administration method may have overestimated the functional status of study participants for most tasks. ICC values indicate that inadequate reliability was found between the 2 assessment methods at 2 weeks; 8 out of 10 tasks had ICC values less than 0.8.

    Table 5 also shows the MBI scores for telephone follow-up and home visit assessments at 3-month follow-up. In general, MBI scores for telephone follow-up were slightly higher than those for home visit assessment, but the only significant differences were found for 4 tasks: feeding, grooming, bathing, and stair climbing. Eight tasks had ICC values less than 0.8, indicating that inadequate reliability was found between the 2 assessment methods at 3 months. Compared with the results of the 2-week follow-up, ICC values showed a general downward trend at the 3-month follow-up.

    Table 5. Comparison of Modified Barthel Index scores evaluated in telephone and home visit assessments at 2-week follow-up (n=28) and 3-month follow-up (n=24).
    VariablesTelephone, mean (SD)Home visit, mean (SD)Z testP valueIntraclass correlation coefficient
    2-week follow-up

    		Feeding8.39 (1.85)6.79 (1.99)−3.22<.0010.66

    		Grooming3.68 (0.72)3.36 (0.78)−1.97.050.58

    		Dressing6.82 (2.00)5.71 (2.02)−2.52.010.64

    		Bathing3.07 (1.15)2.17 (1.19)-3.15<.0010.65

    		Toilet use7.53 (2.11)5.25 (2.19)−3.63<.0010.62

    		Bowels8.54 (1.57)8.32 (1.83)−1.03.310.82

    		Bladder8.86 (1.80)8.82 (1.80)−0.30.760.80

    		Transfer10.60 (2.47)8.64 (3.23)−3.07<.0010.68

    		Mobility10.93 (2.97)8.36 (3.23)−3.21<.0010.63

    		Stairs7.29 (1.88)5.61 (1.79)−3.31<.0010.61
    3-month follow-up

    		Feeding9.00 (1.53)7.37 (2.10)−3.04<.0010.62

    		Grooming4.46 (0.72)4.12 (0.61)−2.20.050.52

    		Dressing7.58 (2.22)6.83 (2.18)−1.50.130.63

    		Bathing3.50 (0.98)2.92 (1.02)−2.08.040.49

    		Toilet use7.17 (2.01)6.58 (1.89)−1.15.250.58

    		Bowels9.20 (1.28)9.04 (1.30)−0.82.410.82

    		Bladder9.08 (1.53)9.13 (1.53)−0.33.740.81

    		Transfer10.54 (2.57)9.79 (3.44)−1.32.190.66

    		Mobility10.37 (2.93)9.42 (3.09)−1.41.160.62

    		Stairs7.29 (1.83)6.54 (1.79)−1.73.080.49

    Feasibility of Using the Videoconference and Telephone Function for Collecting Follow-Up Data

    As shown in Figure 1 and Table 6, 8 out of 60 (13%) participants in the videoconference follow-up group dropped out, and 3 out of 60 (5%) participants in the telephone follow-up group dropped out at 2-week follow-up. There was no significant difference in completion rates between the 2 groups (Χ21=1.6; P=.21) at 2 weeks. At 3-month follow-up, 3 out of 52 (6%) participants in the videoconference follow-up group dropped out, and 9 out of 57 (16%) participants in the telephone follow-up group dropped out. There was no significant difference in the completion rates between the 2 groups (Χ21=1.86; P=.17) at 3 months.

    Table 6. Completion rates at 2-week and 3-month follow-up assessments.
    Completion ratesVideoconference follow-upTelephone follow-upΧ2 (df)P value
    Completion (2-week follow-up)52571.60 (1).21
    Dropout (2-week follow-up)83N/AaN/A
    Completion (3-month follow-up)49481.86 (1).17
    Dropout (3-month follow-up)39N/AN/A

    aN/A: not applicable.

    Table 7 shows participant ratings of satisfaction, comfort, and confidence with using either the videoconference or telephone call function for follow-up assessments. At 2-week follow-up, the majority of participants were either very satisfied (22/52, 42%) or satisfied (29/52, 56%) with the videoconference function, and either very satisfied (19/57, 33%) or satisfied (34/57, 60%) with the telephone call function. There was no significant difference in the satisfaction levels between the 2 groups (Χ23=2.5; P=.28) at 2 weeks. At 3-month follow-up, participants in the videoconference follow-up group reported higher satisfaction than those in the telephone follow-up group (Χ23=13.9; P=.03). Additionally, most participants were either very comfortable (24/52, 46% at 2 weeks; 26/48, 54% at 3 months) or comfortable (28/52, 54% at 2 weeks; 20/48, 42% at 3 months) with the videoconference function, and most participants were either very comfortable (33/57, 58% at 2 weeks; 21/49, 43% at 3 months) or comfortable (24/57, 42% at 2 weeks; 26/49, 53% at 3 months) with the telephone call function. There were no differences in comfort levels between the 2 groups at 2 weeks (Χ23=1.5; P=.22) or 3 months (Χ23=1.3; P=.52). Regarding participant confidence using the videoconference or telephone function for collecting functional data, participants in the videoconference follow-up group rated higher confidence than those in the telephone follow-up group at 2 weeks (Χ23=6.6; P=.04) and 3 months (Χ23=7.9; P=.04).

    Table 7. Ratings of satisfaction, comfort, and confidence at 2-week and 3-month follow-up assessments.
    RatingsVideoconference follow-up, n (%)Telephone follow-up, n (%)Χ2 (df)P value
    Satisfaction (2-week follow-up)2.54 (3).28

    		Very satisfied22 (42)19 (33)


    		Satisfied29 (56)34 (60)


    		Not very satisfied1 (2)4 (7)


    		Unsatisfied0 (0)0 (0)

    Satisfaction (3-month follow-up)13.9 (3).03

    		Very satisfied30 (61)12 (25)


    		Satisfied17 (35)30 (63)


    		Not very satisfied2 (4)4 (8)


    		Unsatisfied0 (0)2 (4)

    Comfort (2-week follow-up)1.50 (3).22

    		Very comfortable24 (46)33 (58)


    		Comfortable28 (54)24 (42)


    		Not very comfortable0 (0)0 (0)


    		Uncomfortable0 (0)0 (0)

    Comfort (3-month follow-up)1.30 (3).52

    		Very comfortable26 (54)21 (43)


    		Comfortable20 (42)26 (53)


    		Not very comfortable2 (4)2 (4)


    		Uncomfortable0 (0)0 (0)

    Confidence (2-week follow-up)6.68 (3).04

    		Very confident33 (63)24 (42)


    		Confident19 (37)30 (53)


    		Not very confident0 (0)3 (5)


    		Unconfident0 (0)0 (0)

    Confidence (3-month follow-up)7.97 (3).04

    		Very confident31 (63)20 (42)


    		Confident18 (37)23 (48)


    		Not very confident0 (0)2 (4)


    		Unconfident0 (0)3 (6)


    This is one of the first studies to compare the validity and reliability of the videoconference and telephone functions of an mHealth app for collecting functional status data in stroke patients after rehabilitation hospitalization and to examine the feasibility and acceptability of using these modes of administration for data collection. We examined these questions prospectively in a cohort of patients who were discharged from inpatient stroke rehabilitation by comparing videoconference and telephone follow-up assessments, as well as comparing these 2 modes of administration to the home visit assessment as the gold standard.

    Validity and Reliability of the Mobile Health App for Collecting Functional Status Data

    We found that most MBI scores obtained by videoconference administration were slightly lower than those obtained by telephone administration, although these score differences did not achieve statistical significance at either 2-week or 3-month periods. Our findings further revealed that videoconference, but not telephone, administration was as valid and reliable as face-to-face, home visit assessment at both 2-week and 3-month follow-up periods. Home visit assessment is conventionally regarded as one of the best methods for collecting posthospitalization outcome measurement data [37]. Home visit (or home rehabilitation) is recognized as providing greater convenience to patients and families while encouraging therapy to continue to occur within the patient’s home; however, it is less cost-effective [38]. One of the reasons for this is that home therapists can often only visit one patient at a time. A previous rehabilitation study indicated that the mean amount of time allotted to perform assessments in home visits is 1 hour and 57 minutes (SD 19 minutes) [39]. Additional time is needed to travel to the patient’s home, which can be even more time consuming for patients who live at a greater distance. Furthermore, inadequate manpower and financial concerns restrict the implementation of home visit assessments for all patients after hospitalization. Instead, other studies [40] have recommended home-based telemedicine as a viable option in the delivery of poststroke recovery programs, because telemedicine has shown promising results in improving the overall health of stroke patients and in supporting caregivers while being delivered by therapists from a distance. The use of technologies appears to be a novel potential approach for the therapist’s assessment of patient performance in home settings. Our findings concur with this notion that an app-based videoconference can be used to assess the functional performance of stroke patients in home settings. The videoconference function may augment other technology-enabled solutions to provide a means of conducting future clinical trials aimed to evaluate the outcomes of any rehabilitation program implemented in the patient’s home [41,42].

    Our findings revealed that almost all MBI scores obtained by telephone administration were higher than those obtained via home visit. This overestimation of functional scores is particularly obvious at 2-week follow-up; telephone assessment of functional status using the MBI was less reliable compared with the home visit at 2-week follow-up, but reliability did improve at 3-month follow-up. Previous studies have attained strong agreement between telephone and face-to-face assessments [43,44]. Psychometric differences may be partially attributed to the use of diverse scales in different studies. Our study used the MBI, whereas 2 other studies used either the Functional Independence Measure or Modified Rankin Scales to measure poststroke disability. Interestingly, Pietra et al [45] conducted a similar study to compare the validity and reliability of the Barthel Index (BI) administered by telephone compared with face-to-face assessment in patients after stroke. They indicated that telephone assessment with the BI is reliable in comparison with face-to-face assessment. Several possible reasons could explain these differences. First, the measurement tool used in our study was the MBI, which is more rigorous and provides more detailed ratings compared with the BI. Second, the stroke sample in our study consisted of individuals who were discharged from the rehabilitation hospital, whereas the stroke sample in their study consisted of inpatients in the hospital. The use of telephone interview is greatly contingent upon whether patients are cognizant of their self-function. In our study, discharged patients who were living in their home and community settings were more likely to experience their actual function, as they have more opportunities to interact with real-world contextual barriers. This view is further supported by our study findings, wherein we found greater agreement between 3-month telephone and home visit assessments than 2-week telephone and home visit assessments; individuals at the 3-month follow-up point have had more exposure to their real-world environmental barriers and are better able to estimate their functional status.

    Feasibility of the Mobile Health App for Collecting Functional Status Data

    It is noteworthy that our findings indicate that completion rates of both videoconference and telephone assessments were greater than 80% at all follow-up periods. These completion rates are within the acceptable range in clinical studies [46]. Moreover, compared with the telephone assessment, patients reported higher satisfaction with and confidence using the videoconference assessment to measure their functional status. A similar study with the WeChat app for health education also revealed high satisfaction perceived by their participants [47]. All of these results confirm that videoconference assessment of the MBI administered via the WeChat app can serve as an alternative tool to the face-to-face, home visit assessment. Videoconference follow-up provides a surveillance platform for clinicians to objectively assess the task performance of patients in their homes. Patients may also perceive a strong sense of participation, which can improve their psychological condition [48]. Prior research [31] and our results have demonstrated the beneficial effects of the WeChat app as a time-effective, cost-effective, and acceptable communication tool for follow-up data collection. However, implementing routine follow-up measurement via technology involves a number of considerations, including the selection of appropriate patients, settings, timing of assessment, and the optimal mode of administration [49]. Offering different modes of administration, such as video consultation, voice communication, text messaging, or image sharing, may help minimize biased sampling and increase patient participation. Future research may consider adopting a mixed method approach that could help to identify facilitators and barriers to the adoption of mHealth apps for collecting posthospitalization data. In addition, the choice of modalities for monitoring patients depends on the size and structure of the organization. An earlier study [50] found that larger organizations report fewer barriers to using technology-based therapeutic tools, likely due to greater resources. Thus, from a researcher or service provider standpoint, this system-level factor is equally important in determining the best mode of administration for follow-up data collection after discharge from inpatient rehabilitation.

    Study Strengths and Limitations

    Our study has a number of strengths. Prior research testing the feasibility and psychometrics of mHealth assessments has primarily adopted the observational design, but ours is one of the first few studies to employ the randomized controlled design to compare these characteristics for 2 different mHealth methods to collect outcome data after discharge from inpatient stroke rehabilitation. Additionally, this study conducted 2 follow-up sessions for both assessment methods to better understand any issues related to long-term compliance. Yet, our study has several limitations. First, patients were recruited from one rehabilitation hospital in a coastal province in eastern China; therefore, results may not generalize to persons living in other regions. Second, this study only included participants who owned a mobile phone and served as their own informant to complete the assessment; patients without a mobile phone and those with severe cognitive or communication impairments may have been excluded from this study. Third, the MBI was the only outcome assessment used in the study. Future research should explore mHealth assessments for measuring other health outcomes in poststroke individuals after discharge from the hospital. Another limitation includes measuring acceptability through 3 self-constructed items (ie, satisfaction, comfort, and confidence), which are limited in their measurement of this construct and do not provide actionable data with which to inform future research. Even though these self-constructed items allow for efficient quantification of acceptability, future research should add a qualitative or mixed method approach to provide additional interpretation and meaning to the quantitative results [51]. Furthermore, this study did not record the amount of time required for performing videoconference and telephone follow-up assessments. Future research should measure the duration of these 2 follow-up assessments and compare them with the time needed for the home visit assessment to provide additional validation evidence. Future research is also needed to conduct a randomized controlled trial comparing these 2 modes of mobile administration to traditional telephone interview method (eg, landline phone service) for stroke patients as a control condition.

    Conclusions

    This study found satisfactory feasibility and validity of an app-based videoconference method for collecting functional data after inpatient stroke rehabilitation. High completion and acceptability, as well as adequate validity and reliability of the videoconference follow-up method, may support its clinical application in poststroke home rehabilitation programs and long-term health monitoring after hospitalization.

    Acknowledgments

    The contents do not necessarily represent the policy of the funding agencies. The authors certify that all financial and material support for this research and work are clearly identified in the manuscript. The authors thank Megen Devine at Washington University for her editorial assistance with this manuscript. This study was funded by the 12th Five-Year Plan Supporting Project of Ministry of Science and Technology of the People’s Republic of China (2013BAI10B01). It was supported by the Collaborative Innovation Center of Rehabilitation Technology. The US National Institutes of Health (K01HD095388) supported a portion of the senior author AWKW’s effort for developing this manuscript.

    Conflicts of Interest

    None declared.

    ‎
    Multimedia Appendix 1

    CONSORT-eHEALTH checklist (V 1.6.1).

    PDF File (Adobe PDF File), 828 KB
    1. Johnston SC, Mendis S, Mathers CD. Global variation in stroke burden and mortality: estimates from monitoring, surveillance, and modelling. Lancet Neurol 2009 Apr;8(4):345-354. [CrossRef] [Medline]
    2. Wang W, Jiang B, Sun H, Ru X, Sun D, Wang L, NESS-China Investigators. Prevalence, incidence, and mortality of stroke in China: results from a nationwide population-based survey of 480 687 adults. Circulation 2017 Feb 21;135(8):759-771. [CrossRef] [Medline]
    3. Broomfield NM, Quinn TJ, Abdul-Rahim AH, Walters MR, Evans JJ. Depression and anxiety symptoms post-stroke/TIA: prevalence and associations in cross-sectional data from a regional stroke registry. BMC Neurol 2014 Oct 1;14:198 [FREE Full text] [CrossRef] [Medline]
    4. Di Carlo A. Human and economic burden of stroke. Age Ageing 2009 Jan;38(1):4-5. [CrossRef] [Medline]
    5. Ferro JM, Caeiro L, Figueira ML. Neuropsychiatric sequelae of stroke. Nat Rev Neurol 2016 May;12(5):269-280. [CrossRef] [Medline]
    6. Boden-Albala B, Litwak E, Elkind MS, Rundek T, Sacco RL. Social isolation and outcomes post stroke. Neurology 2005 Jun 14;64(11):1888-1892. [CrossRef] [Medline]
    7. Wu S, Wu B, Liu M, Chen Z, Wang W, Anderson CS, China Stroke Study Collaboration. Stroke in China: advances and challenges in epidemiology, prevention, and management. Lancet Neurol 2019 Apr;18(4):394-405. [CrossRef] [Medline]
    8. Centers for Disease ControlPrevention (CDC). Outpatient rehabilitation among stroke survivors--21 States and the District of Columbia, 2005. MMWR Morb Mortal Wkly Rep 2007 May 25;56(20):504-507 [FREE Full text] [Medline]
    9. Zhang L, Yan T, You L, Li K. Barriers to activity and participation for stroke survivors in rural China. Arch Phys Med Rehabil 2015 Jul;96(7):1222-1228. [CrossRef] [Medline]
    10. Koh GC, Yen SC, Tay A, Cheong A, Ng YS, de Silva DA, et al. Singapore Tele-technology Aided Rehabilitation in Stroke (STARS) trial: protocol of a randomized clinical trial on tele-rehabilitation for stroke patients. BMC Neurol 2015 Sep 5;15:161 [FREE Full text] [CrossRef] [Medline]
    11. Levine DA, Galecki AT, Langa KM, Unverzagt FW, Kabeto MU, Giordani B, et al. Trajectory of cognitive decline after incident stroke. J Am Med Assoc 2015 Jul 7;314(1):41-51 [FREE Full text] [CrossRef] [Medline]
    12. Andersen HE, Schultz-Larsen K, Kreiner S, Forchhammer BH, Eriksen K, Brown A. Can readmission after stroke be prevented? Results of a randomized clinical study: a postdischarge follow-up service for stroke survivors. Stroke 2000 May;31(5):1038-1045. [CrossRef] [Medline]
    13. Wu HM, Tang JL, Lin XP, Lau J, Leung PC, Woo J, et al. Acupuncture for stroke rehabilitation. Cochrane Database Syst Rev 2006 Jul 19(3):CD004131. [CrossRef] [Medline]
    14. Kwakkel G, Veerbeek JM, van Wegen EE, Wolf SL. Constraint-induced movement therapy after stroke. Lancet Neurol 2015 Feb;14(2):224-234 [FREE Full text] [CrossRef] [Medline]
    15. McHorney CA, Kosinski M, Ware JE. Comparisons of the costs and quality of norms for the SF-36 health survey collected by mail versus telephone interview: results from a national survey. Med Care 1994 Jun;32(6):551-567. [CrossRef] [Medline]
    16. Faria MC, Mateus CL, Coelho F, Martins R, Barros H. [Mail questionnaires. A useful strategy for the follow-up of patients with a cerebrovascular stroke?]. Acta Med Port 1997 Jan;10(1):61-65 [FREE Full text] [Medline]
    17. Janssen PM, Visser NA, Mees SM, Klijn CJ, Algra A, Rinkel GJ. Comparison of telephone and face-to-face assessment of the modified Rankin Scale. Cerebrovasc Dis 2010 Jan;29(2):137-139. [CrossRef] [Medline]
    18. Chowdhury M, Birns J, Rudd A, Bhalla A. Telemedicine versus face-to-face evaluation in the delivery of thrombolysis for acute ischaemic stroke: a single centre experience. Postgrad Med J 2012 Mar;88(1037):134-137. [CrossRef] [Medline]
    19. Lannin NA, Anderson C, Lim J, Paice K, Price C, Faux S, et al. Telephone follow-up was more expensive but more efficient than postal in a national stroke registry. J Clin Epidemiol 2013 Aug;66(8):896-902. [CrossRef] [Medline]
    20. Savio K, Pietra GL, Oddone E, Reggiani M, Leone MA. Reliability of the modified Rankin Scale applied by telephone. Neurol Int 2013 Feb 11;5(1):e2 [FREE Full text] [CrossRef] [Medline]
    21. Revicki DA, Tohen M, Gyulai L, Thompson C, Pike S, Davis-Vogel A, et al. Telephone versus in-person clinical and health status assessment interviews in patients with bipolar disorder. Harv Rev Psychiatry 1997;5(2):75-81. [CrossRef] [Medline]
    22. Beaver K, Luker KA. Follow-up in breast cancer clinics: reassuring for patients rather than detecting recurrence. Psychooncology 2005 Feb;14(2):94-101. [CrossRef] [Medline]
    23. Goldstein LB, Lyden P, Mathias SD, Colman SS, Pasta DJ, Albers G, et al. Telephone assessment of functioning and well-being following stroke: is it feasible? J Stroke Cerebrovasc Dis 2002;11(2):80-87. [CrossRef] [Medline]
    24. Donoff B, McDonough JE, Riedy CA. Integrating oral and general health care. N Engl J Med 2014 Dec 11;371(24):2247-2249. [CrossRef] [Medline]
    25. Zeng F, Deng G, Wang Z, Liu L. WeChat: a new clinical teaching tool for problem-based learning. Int J Med Educ 2016 Apr 25;7:119-121 [FREE Full text] [CrossRef] [Medline]
    26. Rasmussen RS, Østergaard A, Kjær P, Skerris A, Skou C, Christoffersen J, et al. Stroke rehabilitation at home before and after discharge reduced disability and improved quality of life: a randomised controlled trial. Clin Rehabil 2016 Mar;30(3):225-236. [CrossRef] [Medline]
    27. Statista. Number of Monthly Active WeChat Users From 2nd Quarter 2011 to 3rd Quarter 2019 (in Millions)   URL: https://www.statista.com/statistics/255778/number-of-active-wechat-messenger-accounts/ [accessed 2020-02-27]
    28. Chan WS, Leung AY. Use of social network sites for communication among health professionals: systematic review. J Med Internet Res 2018 Mar 28;20(3):e117 [FREE Full text] [CrossRef] [Medline]
    29. He C, Wu S, Zhao Y, Li Z, Zhang Y, Le J, et al. Social media-promoted weight loss among an occupational population: cohort study using a WeChat mobile phone app-based campaign. J Med Internet Res 2017 Oct 23;19(10):e357 [FREE Full text] [CrossRef] [Medline]
    30. Liu J, Zheng X, Chai S, Lei M, Feng Z, Zhang X, et al. Effects of using WeChat-assisted perioperative care instructions for parents of pediatric patients undergoing day surgery for herniorrhaphy. Patient Educ Couns 2018 Aug;101(8):1433-1438. [CrossRef] [Medline]
    31. Lyu KX, Zhao J, Wang B, Xiong G, Yang W, Liu Q, et al. Smartphone application WeChat for clinical follow-up of discharged patients with head and neck tumors: a randomized controlled trial. Chin Med J (Engl) 2016 Dec 5;129(23):2816-2823 [FREE Full text] [CrossRef] [Medline]
    32. Duncan PW, Jorgensen HS, Wade DT. Outcome measures in acute stroke trials: a systematic review and some recommendations to improve practice. Stroke 2000 Jun;31(6):1429-1438. [CrossRef] [Medline]
    33. Martinsson L, Eksborg S. Activity Index - a complementary ADL scale to the Barthel Index in the acute stage in patients with severe stroke. Cerebrovasc Dis 2006;22(4):231-239. [CrossRef] [Medline]
    34. Hsueh I, Wang C, Sheu C, Hsieh C. Comparison of psychometric properties of three mobility measures for patients with stroke. Stroke 2003 Jul;34(7):1741-1745. [CrossRef] [Medline]
    35. Leung SO, Chan CC, Shah S. Development of a Chinese version of the Modified Barthel Index-- validity and reliability. Clin Rehabil 2007 Oct;21(10):912-922. [CrossRef] [Medline]
    36. Shrout PE. Measurement reliability and agreement in psychiatry. Stat Methods Med Res 1998 Sep;7(3):301-317. [CrossRef] [Medline]
    37. Drummond AE, Whitehead P, Fellows K, Sprigg N, Sampson CJ, Edwards C, et al. Occupational therapy predischarge home visits for patients with a stroke (HOVIS): results of a feasibility randomized controlled trial. Clin Rehabil 2013 May;27(5):387-397. [CrossRef] [Medline]
    38. Chen AW, Koh YT, Leong SW, Ng LW, Lee PS, Koh GC. Post community hospital discharge rehabilitation attendance: self-perceived barriers and participation over time. Ann Acad Med Singapore 2014 Mar;43(3):136-144 [FREE Full text] [Medline]
    39. Einarsson U, Gottberg K, Fredrikson S, Bergendal G, von Koch L, Holmqvist LW. Multiple sclerosis in Stockholm County. A pilot study exploring the feasibility of assessment of impairment, disability and handicap by home visits. Clin Rehabil 2003 May;17(3):294-303. [CrossRef] [Medline]
    40. Schwamm LH, Holloway RG, Amarenco P, Audebert HJ, Bakas T, Chumbler NR, American Heart Association Stroke Council, Interdisciplinary Council on Peripheral Vascular Disease. A review of the evidence for the use of telemedicine within stroke systems of care: a scientific statement from the American Heart Association/American Stroke Association. Stroke 2009 Jul;40(7):2616-2634. [CrossRef] [Medline]
    41. Dobkin BH. A Rehabilitation-Internet-of-Things in the home to augment motor skills and exercise training. Neurorehabil Neural Repair 2017 Mar;31(3):217-227 [FREE Full text] [CrossRef] [Medline]
    42. Dobkin BH, Dorsch A. The promise of mHealth: daily activity monitoring and outcome assessments by wearable sensors. Neurorehabil Neural Repair 2011;25(9):788-798 [FREE Full text] [CrossRef] [Medline]
    43. Korner-Bitensky N, Wood-Dauphinee S. Barthel Index information elicited over the telephone. Is it reliable? Am J Phys Med Rehabil 1995;74(1):9-18. [CrossRef] [Medline]
    44. Shinar D, Gross CR, Bronstein KS, Licata-Gehr EE, Eden DT, Cabrera AR, et al. Reliability of the activities of daily living scale and its use in telephone interview. Arch Phys Med Rehabil 1987 Oct;68(10):723-728. [Medline]
    45. Pietra GL, Savio K, Oddone E, Reggiani M, Monaco F, Leone MA. Validity and reliability of the Barthel index administered by telephone. Stroke 2011 Jul;42(7):2077-2079. [CrossRef] [Medline]
    46. Richardson PE. David Sackett and the birth of evidence based medicine: how to practice and teach EBM. Br Med J 2015 Jun 8;350:h3089. [CrossRef] [Medline]
    47. Li W, Han LQ, Guo YJ, Sun J. Using WeChat official accounts to improve malaria health literacy among Chinese expatriates in Niger: an intervention study. Malar J 2016 Nov 24;15(1):567 [FREE Full text] [CrossRef] [Medline]
    48. Cao Y, Lin SH, Zhu D, Xu F, Chen Z, Shen H, et al. WeChat public account use improves clinical control of cough-variant asthma: a randomized controlled trial. Med Sci Monit 2018 Mar 14;24:1524-1532 [FREE Full text] [CrossRef] [Medline]
    49. Wong AW, Heinemann AW, Miskovic A, Semik P, Snyder TM. Feasibility of computerized adaptive testing for collection of patient-reported outcomes after inpatient rehabilitation. Arch Phys Med Rehabil 2014 May;95(5):882-891. [CrossRef] [Medline]
    50. Ramsey A, Lord S, Torrey J, Marsch L, Lardiere M. Paving the way to successful implementation: identifying key barriers to use of technology-based therapeutic tools for behavioral health care. J Behav Health Serv Res 2016 Jan;43(1):54-70 [FREE Full text] [CrossRef] [Medline]
    51. Ramsey AT, Wetherell JL, Depp C, Dixon D, Lenze E. Feasibility and acceptability of smartphone assessment in older adults with cognitive and emotional difficulties. J Technol Hum Serv 2016;34(2):209-223 [FREE Full text] [CrossRef] [Medline]

    ‎
    ADLs: activities of daily living
    BI: Barthel Index
    ICC: intraclass correlation coefficient
    MBI: Modified Barthel Index
    mHealth: mobile health

    Edited by G Eysenbach; submitted 28.11.19; peer-reviewed by Y Guo, L Wang; comments to author 18.12.19; revised version received 08.01.20; accepted 07.02.20; published 13.05.20

    Copyright

    ©Li Li, Jia Huang, Jingsong Wu, Cai Jiang, Shanjia Chen, Guanli Xie, Jinxin Ren, Jing Tao, Chetwyn C H Chan, Lidian Chen, Alex W K Wong. Originally published in JMIR mHealth and uHealth (http://mhealth.jmir.org), 13.05.2020.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mHealth and uHealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.