Mobile tablet-based therapies following stroke: A systematic scoping review of administrative methods and patient experiences

Background and purpose Stroke survivors are often left with deficits requiring rehabilitation to recover function and yet, many are unable to access rehabilitative therapies. Mobile tablet-based therapies (MTBTs) may be a resource-efficient means of improving access to timely rehabilitation. It is unclear what MTBTs have been attempted following stroke, how they were administered, and how patients experienced the therapies. The review summarizes studies of MTBTs following stroke in terms of administrative methods and patient experiences to inform treatment feasibility. Methods Articles were eligible if they reported the results of an MTBT attempted with stroke participants. Six research databases were searched along with grey literature sources, trial registries, and article references. Intervention administration details and patient experiences were summarized. Results The search returned 903 articles of which 23 were eligible for inclusion. Most studies were small, observational, and enrolled chronic stroke patients. Interventions commonly targeted communication, cognition, or fine-motor skills. Therapies tended to be personalized based on patient deficits using commercially available applications. The complexity of therapy instructions, fine-motor requirements, and unreliability of internet or cellular connections were identified as common barriers to tablet-based care. Conclusions Stroke patients responded positively to MTBTs in both the inpatient and home settings. However, some support from therapists or caregivers may be required for patients to overcome barriers to care. Feasibility studies should continue to identify the administrative methods that minimize barriers to care and maximize patient adherence to prescribed therapy regiments.

barriers to care. Feasibility studies should continue to identify the administrative methods that minimize barriers to care and maximize patient adherence to prescribed therapy regiments.

Rationale
Stroke survivors experience many post-stroke complications limiting their ability to function independently, including limitations to mobility [1], upper-limb function [2], communication [3], and cognition [4]. Specialized stroke rehabilitation has been shown to improve functional independence better than non-specialized care [5], and is most effective when performed early [6,7] and intensely post-stroke [8]. However, due to the growing number of stroke survivors [9] and lack of rehabilitation resources [10][11][12][13], many patients are not able to begin rehabilitative therapies early in order to maximize functional recovery. Easily accessible and resourceefficient stroke rehabilitation is needed to provide all stroke survivors with the opportunity to recover their functional abilities and improve their independence.
Mobile tablet computers with therapeutic applications could potentially offer a means of providing early and resource-efficient stroke rehabilitation. Tablet computers can be easily purchased and there exist many inexpensive or free therapy applications with scenarios analogous to those used in stroke rehabilitation. These devices are portable with fairly large, touchresponsive screens, which could be manipulated by stroke survivors depending on their poststroke disabilities. Rehabilitation therapists can prescribe and monitor specific applications to patients based on their post-stroke difficulties.
Mobile tablet-based therapies (MTBTs) following stroke are an exciting new paradigm with many possibilities. However much remains unclear about this approach including the suitable population, the appropriate timing and setting, the necessary patient-support structure, and general therapy adherence. To better understand the feasibility of this intervention, it is critical to understand previous approaches to MTBT administration and the resulting patient experience, including treatment barriers and therapy adherence. A thorough review of the literature may provide important information for the successful conduct of large and potentially costly randomized controlled studies designed to demonstrate treatment efficacy.

Objective
The objective of this study was to summarize the administration of mobile tablet-based therapy (MTBTs) following stroke and the subsequent patient experience to inform treatment feasibility. In particular, we sought to accomplish this objective by answering three research questions: 1. What post-stroke deficits or complications have been targeted by MTBTs and how were the therapies administered?
2. What barriers to care did patients experience while engaging in MTBTs following stroke?
3. Were patients adherent to MTBTs following stroke?

Protocol and registration
The design of this scoping review was guided by the most current and widely known guide for scoping reviews [14], and by PRISMA-P [15] and PRISMA where applicable [16]. PROSPERO currently does not accept scoping review protocols for registration and therefore the study protocol was hosted on the University of Ottawa research depository and can be accessed here: http://hdl.handle.net/10393/35696 (S1 Protocol).

Eligibility criteria
All included articles were required to meet the following criteria: (1) the study population includes adult stroke survivors (18 years or older) of any type (ischemic/hemorrhagic) or stage (acute/chronic) in any setting, and (2) the study intervention involves stroke survivors interacting with a mobile tablet in response to a post-stroke deficit or complication. Articles were excluded if they met one or more of the following criteria: (1) the mobile tablet is primarily used by someone other than the stroke survivor, (2) the mobile tablet is more correctly described as an E-reader, and (3) the manuscript is a study protocol or conference abstract containing data otherwise available from a full study manuscript. We defined MTBTs as patient-driven interactions with mobile tablets via various modalities for therapeutic purposes in response to a deficit, complication, or in order to prevent further health deterioration. There were no restrictions with regards to comparators, outcomes, study design and context, or settings, and conference abstracts were included if no full study manuscript could be found. However, only articles written in English were included and the search was limited to articles written from 2010 onwards as mobile tablets did not become widely popular until this time.

Information sources
Six databases were searched: MEDLINE (OVID interface), EMBASE (OVID interface), Psy-cINFO (OVID interface), CINAHL, Cochrane Database, and Web of Science. Additional sources were used to augment the database search for academic material: (1) a snowball search of relevant articles and reviews identified by the database search, (2) stroke research-related organizational websites, and (3)

Study selection
A two-stage screening process performed by two independent reviewers was used to select included studies. In stage one, two authors (MP, DJ) screened abstracts and titles for potentially relevant articles, and in stage two, two authors (MP, DJ) read potentially relevant articles to confirm they met the inclusion criteria.

Data collection and data items
Two authors (MP, DJ) independently extracted data with the assistance of a data extraction form (S2 Appendix). Data items were selected based on the research objective and questions stated above. Items included general study information, participant characteristics, intervention details, comparator description, study outcome description and results, study setting and other contextual information.

Risk of bias in individual studies
As per current guidelines, no risk of bias assessment was performed for individual studies [14].

Analyses
The review goals and content were not appropriate for quantitative synthesis techniques commonly used in meta-analyses of systematic reviews. However, general study characteristics, participants, interventions, comparators, and patient-reported experiences were summarized narratively and using descriptive statistics where appropriate.

Study selection
The search returned 868 articles from database searches, 8 potentially relevant articles from grey literature searches, 11 unique clinical trials from registry searches, and 16 potentially relevant articles through snowball reference searches of included studies for a total of 903 articles (Fig 1). Title and abstract screening narrowed the search down to 60 articles of which 37 were excluded leaving 23 articles for inclusion . The eligible articles came from various search sources; 16 from database searches, 2 from grey literature, 2 from snowball searches, and included both full-texts and abstracts. Seven of the included articles were abstracts, the remaining articles were full manuscripts. None of the clinical trials identified by registry searches provided preliminary data or were completed, and therefore were not formally included.

Study characteristics
Most of the included studies were observational; 10 cohort studies, 3 case studies, 2 cross-sectional studies, and 1 qualitative interview (Table 1). Of the 7 experimental studies, 3 were RCTs, 2 were randomized experiments, 1 was a non-randomized experiment, and 1 used a cross-over design. Sample sizes were small ranging from 1 to 63 with an average of 18 participants. Three articles  (1), Salaris et al. 2016 (2), and Janssen et al. 2016) were conference abstracts based off the same "TnT" study, although they reported different results.
The majority of studies recruited participants in the chronic or subacute stages of stroke (14 studies), and only 2 studies included acute stroke patients. Participant stroke stage was either not reported or unclear in the remaining 7 studies. Only 8 studies explicitly reported the number of participants who had previous experience with tablets. On average, nearly half of the participants in these studies had previously used a mobile tablet.  (although 3 of these were from the same "TnT" study), 3 for cognitive deficits, 2 for deficits targeted by physiotherapy and 1 each for balance and upper extremity function. Therapies tended to be administered using iPad tablets to provide personalized therapy using a unique assortment of commercially available apps for individual participants (Table 2). Therapies were performed in home or inpatient settings and mostly performed independently with no reported assistance from caregivers or clinicians. However, even when therapies were independent, regular contact was often kept with patients through video-conferencing (Katalinic et

Barriers to tablet-based care following stroke
Barriers to care, methodological challenges, and patient experience were defined as separate outcomes at the protocol stage but were found to be intertwined during data collection (i.e. barriers to care are often patient reported experiences that lead to methodological challenges). Eight studies reported barriers to MTBT care, which were categorized as device, patient and system barriers ( Table 3). The difficulty of the assigned tasks were identified as a device barrier  Kurland et al. 2014). In each case, participants found the assigned tasks too easy to perform. Bugs in provided videoconferencing software (Katalinic et al. 2013) and the high speed of task prompts and requests (White et al. 2015) were also identified as device barriers. However, the software bug was later resolved. The most commonly reported patient barrier was a difficulty in following complex instructions (Kurland et  Finally, one study reported that the patient's internet bandwidth made videoconferencing software unusable (Katalinic et al. 2013) and another reported that a participant was particularly excited to use their device for non-therapeutic purposes (Kurland et al. 2014). In the latter case however, this did not appear to interfere with their therapy progress.
System barriers were mostly limited to a study conducted in the acute setting (Mallet et al. 2016 (1)), however two studies of home-based MTBTs also noted barriers in this domain (Katalinic et al. 2013;White et al. 2015). All three of these studies reported connection issues as interfering with care. Some participants lived in areas with poor 3G coverage, resulting in subpar videoconferencing quality (Katalinic et al. 2013). This was resolved by calling participants to help them connect the devices to their home internet connection. However, some participants in another home study reported finding their home internet connections to be unreliable (White et al. 2015). In addition to connection problems, Mallet et al. 2016 (1) also reported barriers related to legitimate concerns from Hospital Infection Control and Hospital IT Security and challenges related to standard of care assessment procedures. Infection control required tablets be cleaned with disinfecting wipes before being transferred to another patient to prevent the spread of disease. There were concerns about hospital network security requiring an independent security assessment before the study could continue. Finally, five eligible patients did not undergo standard of care assessments for unspecified reasons and were missed for recruitment.
Despite these barriers the overall reported patient response was very positive with many studies reporting high satisfaction and usability through either surveys (

Patient adherence to tablet-based care following stroke
The majority of studies (17/23) prescribed a MTBT dose with only 6 studies not assigning specific MTBT regiments or not reporting this information (Table 4)  Tablet-based therapy usage habits were reported by a minority of studies (Table 5). Tablet usage ranged from little daily usage (Kurland et al. 2015) to daily use equal to or exceeding 1 hour   (1) (2)). Specific tablet usage was not reported by a handful of studies. In particular, Stark & Warburton (2016) could not confirm usage habits using information transmitted from the tablets and relied on patient selfreported usage. White et al. (2015) qualitatively reported the usage habits of a handful of participants using quotes to communicate the variety of usage habits they observed in their participants.

Discussion
This systematic scoping review yielded 23 eligible articles pertaining to mobile tablet-based therapies for a variety of post-stroke deficits and complications. The majority of the identified articles reported the results of observational studies with small samples of chronic or subacute stroke patients. The goals of most of the studies were exploratory, with only a few articles reporting the results of experimental studies attempting to demonstrate treatment effectiveness. The collected body of literature suggests MTBTs following stroke may be feasible and acceptable, with many studies expressing positive experiences. However, common barriers to care were identified that may prevent certain patients from successfully engaging in therapy.

Mobile tablet-based communication, cognitive, and fine-motor therapy following stroke
The majority of the attempted MTBTs targeted communication, cognitive, and fine-motor deficits. Considering interventions for these deficits are arguably the most intuitive to implement on a tablet device, it is not entirely surprising that research has focused mainly on these interventions. Furthermore, there is already research supporting the effectiveness of computer-based speech-language therapy [4]. This research-base, coupled with the emergence of popular therapeutic apps like Constant Therapy, Tactus Therapy, and Dexteria may explain why many researchers are more interested in attempting MTBTs for these deficits over other post-stroke complications. Although the methods of administration varied among MTBTs for communication and fine-motor deficits, some trends emerged.
Most MTBTs were performed independently without clinician guidance, although in some cases caregivers or family members participated in therapy with patients. Regular contact with clinicians through videoconferencing and in-person visits were common even among independently performed therapies. The involvement of others in therapy may be beneficial for patient barriers related to therapy and device complexity could be overcome with assistance. Mobile tablet-based therapies following stroke Contact with clinicians offers an opportunity for patients to voice issues regarding tablet complexity and for therapists to re-explain complex therapy tasks or features to participants and their caregivers. However, therapists should try not to infringe upon patient independence as this was found to be one of the aspects of tablet-based therapy that patients enjoyed. Similarly, patients would likely prefer for physicians to keep contact in a manner that does not require them to leave their home as the home-based aspect of therapy was also frequently noted as a positive attribute. Videoconferencing was successfully used to contact patients, although some technical difficulties were noted and it is possible that not all patients will have access to reliable cellular or wireless internet connections. Mobile tablet-based therapies tended to use personalized tasks tailored to patient deficits. Personalization may prevent patients from encountering therapies with inappropriate difficulty levels and encourage extended therapy engagement. However, this relationship between personalization and therapy engagement was not made clear from the collected usage data. It should be noted that even studies with personalized therapy tasks reported barriers to care involving task difficulty suggesting personalization alone is not enough. Rather, therapies should both be personalized and then adjusted as patients engage in therapy. The result is an iterative process whereby patients are assigned tasks based upon their deficits, engage in therapy, express difficulty or make progress, and then the process begins again as new tasks are assigned. This also further highlights the importance of maintaining regular contact with clinicians who are needed to make informed therapy adjustments.

Mobile tablet-based physiotherapy
Mobility and upper extremity deficits are very common post-stroke [1][2], and yet there were only four studies involving tablet-based physiotherapy activities, two of which were reported as short conference abstracts. Perhaps, the lack of physiotherapy interventions reflects the difficulty of translating traditional physiotherapy to tablets or perhaps due to interests in other promising technology such as robotics [40,41]. Although the few studies reported heterogeneous administrative methods, interesting ideas for tablet-based physiotherapy were described. One intervention for upper extremity therapy used a combination of tablet sensors with a smartphone attached to the patient upper limb to track movement in response to tablet therapy activities which was well-received by patients (Choi et al. 2016 (2)). Other studies used video capabilities to provide visual feedback on movements or provide visual demonstrations (Hiyamizu et al. 2013), and assigned patients a Fitbit to track activity levels (Crotty et al. 2014).
The ability of tablets to host video conferences with therapists could be particularly useful for providing tablet-based physiotherapy. Crotty et al. 2013 reported using a combination of video conferences to deliver therapy, and video recording to save sessions for later viewing, however the details of this administrative method with regards to physiotherapy were not made clear. Further detailed studied of similar administrative approaches and barriers to care would be helpful for determining the feasibility of tablet-based physiotherapy in the home setting. Although patients would likely enjoy the convenience of home physiotherapy instead of travelling to or staying in a rehabilitation center, the safety of home-based physiotherapy is unknown.

Mobile tablet-based depression treatment
There were no tablet-based mood interventions published, despite depression being common following stroke [42]. However, three abstracts from one study [26,36,37] and one further study [29] reported on interventions targeting quality of life, a construct related to depression. Although the results provide some promise that simply providing patients with a tablet computer could improve quality of life, and therefore mood, it remains unknown if a more structured approach to tablet-based mood therapy is feasible. Interestingly, another study reported improvements in mood and arm function after a physiotherapy intervention [33]. This suggests that providing tablet-based interventions to improve physical deficits could indirectly lead to improvements in post-stroke depression. Logically some patients would feel depressed following the sudden loss of their physical abilities and independence following stroke, and it follows that improvements in physical function could be coupled with an improvement in mood. Future studies of tablet-based intervention targeting physical abilities should consider measuring mood as a secondary outcome to further explore this hypothesis.
Designing and administering tablet-based interventions directly targeting mood is difficult, as no intervention has distinguished itself as particularly effective with stroke patients using a traditional face-to-face administrative approach. For example, although there has been considerable research on internet-based cognitive behavioral therapy in other populations [43,44], it remains unclear if this approach is effective for individuals post-stroke [45]. However, there has been promising evidence for the effectiveness of mindfulness-based therapies for improving mood post-stroke [46] and there exists the possibility of translating these interventions to tablet devices in the future.

Mobile tablet-based therapy for acute stroke patients
Early rehabilitation initiation is important to maximize recovery of function [6,7], and MTBTs could be used to provide early therapy during the significant amount of downtime experienced by patients in acute care [11]. Despite this, the only two studies attempting MTBTs in the acute stroke setting to provide early stroke rehabilitation services (Mallet et al. 2016 (1); Mallet et al. 2016 (2)) were performed by our own Ottawa Stroke Program Research Team. The technology, need, and available time exist to administer tablet-based therapy in the acute stroke setting. Further investigation into this area is warranted.

The next steps for mobile tablet-based therapies following stroke
This review provides a summary of MTBT administration methods, barriers to care, and therapy adherence following stroke. The accumulated studies suggest many stroke patients are able to successfully engage in tablet-based therapies for communication, cognitive and finemotor deficits with minimal assistance and enjoy the independence and convenience of tabletbased therapy despite barriers to care. Attempting randomized controlled trials for therapies targeting communication, cognitive, and fine-motor deficits may not yet be appropriate despite seemingly strong support for treatment feasibility. Further studies should track therapy adherence and experiment with methods for encouraging patient adherence to therapy regiments. Successful methods for promoting therapy engagement may be simple as sending text notifications to the tablet device, setting weekly therapy goals, regular videoconferences, or perhaps a combination of strategies individually tailored to patient preferences.

Limitations
The large breadth of outcome information eligible for extraction from articles by the two independent reviewers means some relevant outcomes could have been missed. However, the data extraction process used assistive documents to help reviewers identify relevant outcomes. No risk of bias assessment was performed on the included studies as is common for scoping reviews. However, this was done in accordance with current scoping review guidelines and considering the goals of the review were to collect information on administrative methods and barriers to care, a risk of bias assessment would be unlikely to impact our findings.

Conclusions
The majority of MTBTs following stroke targeted communication, cognitive, and fine-motor deficits, and were positively received by patients despite barriers to care. The findings suggest tablet-based therapy may be feasible for certain stroke deficits although little is known about therapy adherence. Feasibility studies should continue to refine the administrative methods for frequently targeted post-stroke deficits to minimize barriers to care and maximize treatment adherence.