Aphasia arising post-stroke is an acquired communication disorder characterized by impairment in linguistic abilities. It stems from varying degrees of damage to the language center of the brain (usually located in the left hemisphere), which affects oral expression, reading ability, writing ability, language comprehension, and even cognitive and computational functions (1, 2). After an initial ischemic stroke, ~30% of patients may manifest symptoms of aphasia (3). Studies reveal that during the year following a stroke, ~43% of individuals with aphasia still confront ongoing communication challenges (4). Aphasia significantly hinders daily life functioning, subsequently reducing quality of life and potentially leading to issues such as depression (5–8).

In recent years, an increasing number of studies have focused on the use of mobile applications, such as computers and tablets, to enhance the expressive language skills of individuals with aphasia (9, 10). For example, Zhou et al. (11) found that a 14-day, 30-min-a-day computer-based training session resulted in more significant speech improvements compared to traditional treatments. Mobile application-based therapy is gaining attention as a means of remote delivery (12). One potential advantage of this treatment modality is its potential for increased cost-effectiveness, reduced therapist burden, and enhanced patient satisfaction and treatment adherence (13).

Three systematic review studies have already explored the impacts of various innovative technologies on aphasia, and their consistent findings suggest that innovative technologies hold promise in improving language functions among individuals with aphasia (14–16). However, the systematic reviews conducted by Lavoie et al. (14) and Russo et al. (16) focused only on the impact of a specific technology on language performance in a particular domain for people with aphasia and the literature included was limited to studies conducted prior to 2017. Additionally, the review by Repetto et al. covered literature from only three databases (15). Hence, the objective of this present systematic review is to comprehensively explore the impact of mobile application-based speech-language therapy (SLT) on language functional performance across multiple domains in individuals with aphasia.

Interventions based on mobile applications have gained increasing popularity for individuals with aphasia, primarily utilizing computers and tablets. These electronic devices have become integral to people's lives and have also brought convenience to rehabilitation therapy. Among the studies included, 66% allowed aphasic patients to use software for training at home, while therapist involvement during clinic-based training was primarily focused on addressing technical issues rather than guiding the therapy process. This underscores the potential for individuals with aphasia to engage in speech-language therapy either independently or with minimal assistance. A systematic review of 10 randomized controlled trials, four pre-post controlled studies, and one crossover trial indicates a high risk of bias. Heterogeneity of applications, varied outcome measures, differing intervention intensities, variations in aphasia onset times and severity, result in diverse study outcomes. This makes it challenging to assess and conduct meta-analyses across studies (36). Most of the studies had selection bias, all randomized controlled trials used random allocation, while five of them did not specify the exact method of randomization (18, 20, 23, 25, 26), leading to uncertain risks, and 70% of studies did not conceal allocation. There were no studies describing blinding of subjects, thus introducing performance bias. Monitoring bias had a better performance with 8 studies reporting assessor blinding, but 2 did not describe blinding procedures (18, 25). The five non-randomized controlled studies showed significant bias, lacking random allocation (27–31). Furthermore, according to the TREND report, none of the 5 studies described blinding. Thus, overall high bias across all studies is unfavorable for determining the effectiveness of mobile application-based speech-language therapy. However, overall, mobile application-based SLT holds significant promise and potential for improving the performance of individuals with aphasia.

The results of the study revealed that eight studies indicated the effectiveness of mobile application-based speech-language therapy (SLT) in improving the overall language functioning of individuals with aphasia. Of these, four studies showed no significant difference in improvement between the intervention and control groups. Mobile application-based therapy shows promise in enhancing language function; however, it is uncertain whether it is superior to the effects of traditional therapy. There were inconsistent results from five studies regarding whether mobile application-based therapy could produce transferable effects in terms of communicative and expressive language skills.

Four of the 10 RCTs compared mobile application-based speech-language therapy with Speech therapist (ST)-mediated therapy (19–21, 23). These studies aimed to explore whether computer-based interventions were comparable or potentially superior to face-to-face ST interventions. Among these, 3 studies found that computer-based interventions had outcomes for aphasic patients that were either better than or non-inferior to those of ST-mediated treatment. However, a study by Kesav et al. (20) reported that a regimen of three sessions per week for 60 min of ST treatment was more effective than a schedule of three sessions per week for 120 min of computer-based treatment. This discrepancy could potentially be attributed to the relative unfamiliarity of the included aphasic patients with computer programs (20). The study by Kesav et al. (20) did not provide detailed descriptions of the feedback mechanisms and patient compliance related to computer-based treatment. Personalized therapy and feedback are crucial factors for enhancing treatment efficacy (37). However, in our study, only 40% of the reports mentioned the feedback methods used in mobile application-based interventions, and these feedback mechanisms typically displayed errors made by patients during training. This type of feedback is relatively limited compared to the feedback provided by ST. Future research could explore whether more diverse feedback methods might have an impact on treatment outcomes.

Treatments that can be delivered at home can reduce the financial burden on patients and provide a higher intensity of treatment. Several studies have emphasized the need for intensive treatment programs for aphasia (38, 39). According to a network meta-analysis, the greatest gains in overall language proficiency were associated with >20 h of SLT (40). Breitenstein et al. observed that engaging in ≥10 h of intensive speech-language therapy per week for 3 weeks effectively improved communication skills in patients with chronic aphasia (41). Clearly, delivering therapy through mobile applications enables patients to achieve higher treatment intensity while participating in independent treatment. However, among the studies we included, only one managed to reach a treatment intensity of ≥10 h per week. Cherney et al. (26) founded that engaging in computer-based language imitation therapy six times per week, with each session lasting 90 min, effectively enhanced the overall language abilities of the patients. However, the training effects did not show significant differences when compared to the control group. This could potentially be attributed to the fact that the control group utilized computer games designed for memory and attention training. Training focused on cognitive functions might also lead to improvements in language abilities among individuals with aphasia (42). Therefore, it remains uncertain whether intensive, application-based SLT effectively enhances language function in individuals with aphasia. Additionally, application-based interventions offer a cost-effective supplemental approach that can address the limitations of high-intensity, in-clinic speech therapy for individuals who face challenges in accessing such treatment (43).

From the studies we included, the interventions can be divided into two main types: single-component interventions and multi-component interventions. Single-component interventions primarily focus on naming abilities. For instance, Palmer et al. (22) found that computer-based word retrieval training effectively improved naming abilities in individuals with aphasia, but did not significantly enhance their communicative skills. This might be attributed to the narrow focus of the training content. In contrast, Spaccavento et al. (23) employed a computer-based multi-component intervention, which resulted in significant improvements in the communicative abilities of aphasic patients post-treatment. However, as there have been no studies directly comparing the effectiveness of these two types of interventions, we cannot conclude whether there is a difference in the improvement effects between single-component and multi-component interventions. Additionally, the extent to which training effects can transfer to other functional aspects requires further investigation. Similar to findings in studies related to post-stroke motor function, where motor training improved motor skills but had limited effects on overall quality of life and daily functioning, it remains important to determine the potential for transfer effects in aphasia interventions (44, 45). In conclusion, further research is needed to delve into the differences in improvement effects between single-component and multi-component interventions, in order to better inform the application of mobile application-based speech-language therapy in clinical practice.

The sample sizes of the included studies were generally small, ranging from 7 to 169 individuals, with ~66% of the studies having a sample size of <30 individuals. This small sample size may be related to the difficulty of recruiting people with aphasia. However, smaller sample sizes may affect the assessment of treatment effects. In addition, there was variability in the duration of post-stroke in aphasia patients in the included studies, with ~73% of the studies including patients with chronic aphasia, so there is uncertainty about the benefit of treatment for patients with different stages of aphasia. Another challenge was the wide variation in the intervention components used in the included studies, which made it difficult to compare and synthesize the findings. Finally, only eight studies explored the effect of mobile application-based intervention methods on the maintenance of efficacy, with follow-up times ranging from 2 weeks to 6 months, with an average of approximately 12 weeks. However, this follow-up period may be too short to effectively assess whether intervention approaches are able to maintain language functioning in people with aphasia over an extended period of time.

With regard to the many shortcomings and deficiencies of CSLT research, based on these studies, future researchers can develop a more rigorous and standardized procedure to validate the significance of CSLT beyond traditional face-to-face speech therapy. In this procedure, several aspects need consideration: the types and severity of aphasia; the optimal treatment stages of aphasia; maintenance therapy in chronic phases; the best monitoring of naming, perception, communication performance, social participation, and wellbeing; the definition of appropriate control groups; the efficacy and maintenance of intensified therapy; and the management of patient attention and feedback control interference. It is worth noting that the cost-effectiveness of CSLT is not publicly available and requires separate investigation.

The findings of this systematic review suggest that mobile application-based interventions for aphasia hold promise in improving overall language function. However, uncertainties remain regarding the improvement in functional communication abilities and whether gains in naming abilities can transfer to untrained objects. The efficacy of single-component interventions cannot be directly compared to multi-component interventions. Nonetheless, overall, mobile application-based interventions show positive prospects for enhancing speech-language function in aphasia.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number (s) can be found in the article/Supplementary material.

ZJ: Writing – original draft, Software, Methodology, Investigation, Formal Analysis, Data curation, Conceptualization. MH: Writing – review & editing, Supervision, Investigation. CZ: Writing – review & editing, Supervision, Software, Investigation. XC: Writing – review & editing, Visualization, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.