This study adopted a quasi-experimental design. The TYB participants were recruited from a community-based brain health for stroke recovery study. These participants were initially recruited from various service providers in Singapore, including day rehabilitation centers, primary healthcare polyclinics, and stroke associations, as well as through word of mouth. There were 64 participants of the brain health study who met the eligibility criteria and were provided with information about this study; 56 of the 64 (87.5%) consented to participate. The criteria for participant eligibility were either: (1) stroke survivors who were at least 3 months post-stroke and who did not have significant aphasia or apraxia that would prevent them from providing informed consent, or (2) any caregivers of stroke survivors. An additional criterion for all participants was sufficient fluency in English to provide informed consent and participate in the group program.

Despite the multicultural nature of Singapore where individuals are proficient in multiple languages, the English language holds a dominant position in the daily lives of its residents. A landscape study conducted on public places in Singapore indicated that English was frequently employed as the primary language of instruction (20). Consequently, considering the study being a pilot, the researchers opted to recruit participants who possessed a good command of the English language.

Before commencement of the program, study team members sent a survey link to the participants and did home visits to obtain baseline assessments. Participants were then invited to attend the program online, conducted via Zoom video conferencing software application. Once the participants completed the program, the study team conducted home visits again for post-intervention assessments. This study obtained institutional ethics approval from the National University of Singapore-Institutional Review Board (NUS-IRB) (NUS-IRB-2021-412).

The TYB program was conducted in English and facilitated by neuropsychologists, with the content based on an earlier version conducted within a public hospital setting (19). The program was delivered over nine sessions with a duration of 60–90 min per session, with two to three sessions scheduled each week. The program design was tailored to align with the cultural context prevalent in Singapore, which predominantly reflects Asian cultural influences.

One of the facilitators, YD, possessed a professional background as a neuropsychologist with 7 years of experience. During YD’s doctoral studies, she conceptualized the Train-Your-Brain (TYB) cognitive health program, which was subsequently employed on older adults in Singapore (19). Another facilitator, NC, was likewise a neuropsychologist with 6 years of professional experience. YD personally guided NC in the implementation of the TYB program to ensure its proper execution. To uphold program fidelity, YD and XT, a health promotion specialist, developed a comprehensive manual outlining the precise execution of each session. YD prioritized attending the sessions facilitated by NC to ensure adherence to the manual’s guidelines. This ensured that the prescribed steps in the manual were faithfully followed during the sessions. Furthermore, YD offered constructive feedback to NC after each session, highlighting areas that could be improved.

Participants who finished the first eight sessions were deemed to have completed the program. Each session focused on a specific topic related to cognition. Upon conclusion of every session, the participants were assigned homework which entailed practical exercises and activities related to the theme of the session, aimed at reinforcing the concepts learnt. The participants were encouraged to discuss as a group on the strategies that they had adopted. Participants were provided with a link to offer feedback on the session they had attended, which they could access at the conclusion of each session. It is important to note that providing feedback was entirely optional.

The TYB session topics were:

Both stroke survivors and caregivers participated in the same session content as a family unit. While the focus of the program was primarily on meeting the needs of stroke survivors, caregivers were actively encouraged to join the sessions alongside the stroke survivors and engage in practical exercises and activities together. Moreover, caregivers were also encouraged to reinforce the learned concepts by practicing them with their stroke survivors after each session.

The rationale behind the inclusion of caregivers in the TYB sessions stemmed from the utilization of the Theory of Planned Behavior (TPB) proposed by Ajzen (23). This theory primarily centers on individuals who are susceptible to the influence of their social environment, such as friends and family. By employing the TPB, it was anticipated that stroke survivors would be able to observe any modifications in caregivers’ behavior after their participation in the TYB sessions, considering that caregivers are the primary individuals responsible for their care and support. In accordance with the TPB, individuals’ actions or intentions are governed by three fundamental factors: attitude, subjective norms, and perceived control (23).

Attitude refers to an individual’s personal assessment of a health outcome. Attitude consists of two components: experimental and instrumental (23). The experimental component involves the emotional evaluation of a health outcome, whereas the instrumental component encompasses the cognitive evaluation of the expected health outcome and the value attributed to it (23).

The initial session titled “Memory and Thinking Difficulties Related to Stroke” and the sixth session titled “When to Return to Work, Phased Return or Full-time Work?” served as introductory sessions for subsequent topics. The primary objective of the first session was to provide insights into cognitive impairment and its impact on daily functioning. Conversely, the sixth session aimed to shed light on the facilitators and barriers to resuming work. Additionally, these sessions touched upon the positive and negative consequences of both action and inaction in promoting brain health. They also delved deeper into the correlation between pleasure and taking action, particularly in terms of establishing better relationships with their loved ones and caregivers. The ultimate goal of implementing these techniques was to induce behavioral changes among stroke survivors by leveraging the experimental and instrumental components of attitude.

Subjective norms refer to individuals’ beliefs regarding the behavior of their reference groups, encompassing both descriptive norms (beliefs about the behavior exhibited by these groups) and injunctive norms (beliefs about the approval or disapproval of these groups toward specific behaviors) (23). In addition, individuals’ motivation to comply with such behavior stems from their desire to gain the approval of their referent groups.

During sessions 2, 3, 4, 5, 7, and 8, participants were provided with the opportunity to share their experiences and insights regarding the positive outcomes associated with taking action. For instance, in the second session titled “Health and Lifestyle Impact on Memory, Thinking, and Stroke Recovery,” participants were encouraged to discuss the benefits they experienced by incorporating regular exercise into their routines. Furthermore, caregivers were also encouraged to contribute their perspectives. The concluding session in Session 9 solidified the stroke survivors’ beliefs that they were capable of making positive changes based on feedback from group members, and it confirmed the crucial role played by caregivers in supporting their efforts to implement these changes. By employing this technique during these sessions, stroke survivors not only receive encouragement from fellow stroke survivors (descriptive norm), but they also gain knowledge and support from caregivers who are highly supportive of their efforts to make positive changes (injunctive norms).

Perceived control refers to the individual’s belief in their ability to exert control over their behavior. Following each session, participants were assigned homework to practice the strategies taught during the session. The facilitators explicitly instructed participants on how to complete the homework. Caregivers present during the sessions were specifically instructed to encourage stroke survivors to complete the homework and engage in the activities alongside them. Furthermore, at the onset of each session, all participants were encouraged to openly share their personal experiences with others, thereby augmenting the stroke survivors’ perception of control. This collaborative approach aimed to enhance stroke survivors’ confidence in performing positive behaviors.

By employing the TPB, the significance of caregiver participation in the TYB sessions became evident, even though the program was primarily tailored to stroke survivors. Without the involvement of caregivers in the sessions, stroke survivors might lack the intention to make changes in their lives to improve brain health.

Socio-demographic profiles were available from the community-based brain health for stroke recovery study. Before the home visits, participants were asked to complete an online survey which asked about lifestyles, medical adherence, confidence in obtaining employment, quality of life and levels of depression, anxiety, and stress. During the home visits, the participants were evaluated for their cognitive functioning.

The study employed descriptive analysis to examine the demographic profiles and satisfaction levels of the participants. Furthermore, content analysis was utilized to analyze the open-ended feedback provided by the participants.

Given the small sample size (stroke survivors n = 27, and caregivers n = 21) in this study, normality was not assumed. Chi-square Test of Independence was conducted to examine the relationships between categorical variables. Mann–Whitney U-test was conducted to compare outcomes between the groups (stroke survivors and caregivers). Wilcoxon Signed Rank Test was conducted to measure the change in cognitive functioning and mood amongst participants within the same group after TYB. The analyses were conducted using Python, version 3.9.12. A two-tailed value of p of less than 0.05 was considered as statistically significant.

A rigorous methodology was employed to ensure comparability and assess the relative performance of cognitive functioning of the participants. The process involved converting the raw scores of individual tests from the Montreal Cognitive Assessment (MoCA) and Symbol Digit Modalities Test (SDMT) into standardized z-scores, leveraging the means and standard deviations (SDs) specific to each test. Subsequently, the scores from the MoCA and SDMT were combined at the individual level. These combined scores were then averaged, resulting in total z-scores. To derive the cognitive composite scores, the total z-scores were further transformed using the composite mean and SD. By implementing this systematic approach, a comprehensive global cognition score was generated for each participant, enabling a comprehensive evaluation of cognitive functioning.

To further examine the intervention’s effect on depression, anxiety, stress, and cognition among participants, the study conducted a post-hoc individual analysis. The researchers employed Crawford and Howell’s approach (34) to assess whether notable differences in performance existed between the control sample, used as a reference group, and each individual participant. The control sample participants were matched with each individual based on variables such as age, gender, and education level (35). The established norms, derived from a representative sample treated as the parent population, considered the mean and standard deviation as fixed population parameters rather than estimators. This assessment enabled the researchers to determine whether the performance of the tested individual significantly deviated from what would be expected in a typical population. Additionally, the approach provided an accurate estimation of the proportion of individuals in the general population who might achieve lower scores compared to the observed performance of the tested individual (35). This estimation contributed to a comprehensive understanding of how the individual’s performance compared to others within the overall population (34).

To facilitate individual analyses, a computer software program SINGLIMS.EXE was employed, which can be accessed at the following URL: https://homepages.abdn.ac.uk/j.crawford/pages/dept/SingleCaseMethodsComputerPrograms.HTM#conflims. SINGLIMS.EXE has been specifically developed for comparing an individual’s test score with the score of a normative or control sample (36). This approach utilizes a non-central t-test to compare the score of a single case with scores obtained from a control sample. The test determines the extent to which a patient’s score significantly deviates from that of the controls, thereby estimating the abnormality of the patient’s score (i.e., estimating the percentage of the control population with a lower score). Additionally, the test provides associated confidence intervals for this measure. A significant difference of less than 0.05 indicates that the performance of the individual is significantly different from the control sample (35).

Tables 1, 2 display the socio-demographic profiles of the participants: 56.3% were females; 77.1% were of Chinese ethnicity; the median age was 57.0 ± 22.3; and median years of education was 15.0 ± 3.0.

Analysis using Chi-square revealed a significant association between gender and group (stroke survivor; caregiver). More stroke survivors were males, while more caregivers were females [χ² (df = 1) = 4.68, p = 0.03]. Other demographic variables did not reveal any significant association with group.

Eight (14.3%) individuals did not complete the home visit assessments and were therefore excluded from analysis. In total, 48 individuals completed the entire TYB program by successfully attending all eight sessions, thereby meeting the program’s completion criteria. Among these participants, a subgroup of 22.9% (n = 11) took part in Session 9, which was designed as a revision session. Out of the 11 participants who attended this session, 6 were stroke survivors.

The written feedback received from participants were mostly positive. There were 162 recorded responses in the feedback forms. Among the 162 respondents, a high percentage expressed agreement and satisfaction regarding various aspects of the sessions. Specifically, 89.5% (n = 145) found the sessions interesting, while 88.9% (n = 144) indicated that they were easy to understand. A significant majority of 87.0% (n = 141) agreed that the sessions successfully achieved all their aims, and 84.0% (n = 136) stated that they felt confident in applying the strategies taught during the sessions. Furthermore, 89.5% (n = 145) reported enjoying the sessions, and 83.3% (n = 135) found them helpful. Most respondents, 88.9% (n = 144), agreed that the session contents were sufficient, and a substantial 90.7% (n = 147) expressed overall satisfaction with the sessions.

In the study, a total of 20 open-ended feedback responses were collected. Among these, 55.0% (n = 11) indicated that the coping strategies, including the utilization of SMART techniques, mindfulness practices, fatigue and stress management, and stress management, were perceived as helpful. Additionally, 10.0% (n = 2) of the participants mentioned that every aspect of the sessions proved useful, while the remaining 25.0% (n = 7) reported no specific benefits or simply stated “Nil.” In response to inquiries regarding the components of the TYB program that participants did not find valuable, a significant majority of 95.0% (n = 19) indicated “Nil,” indicating no specific areas of dissatisfaction. The sole feedback received highlighted the perceived lack of usefulness of the work-related elements within the TYB program. In relation to the inquiry concerning the depiction of memory and thinking strategies that have positively impacted the participants’ lives, findings indicate that a small portion of the participants (10%, n = 2) employed the use of mnemonic devices by associating names with specific concepts. Additionally, a larger proportion (35%, n = 7) reported the adoption of effective note-taking techniques and the regular practice of setting SMART goals in writing. A minority subset (10%, n = 2) emphasized the significance of maintaining sufficient sleep and exercise routines for cognitive enhancement. A single participant (5%, n = 1) emphasized the importance of cultivating a positive mindset. A subgroup of participants (15%, n = 3) responded with “Nil” when asked about memory and thinking strategies, while the rest responded “Yes.”

A total of 13 verbal responses were documented, from which two themes were derived. Detailed findings derived from the analyzed verbal feedback data are presented in the Supplementary Appendix A.

The second round of home visits took place in December, leading to the formulation of the first theme titled “Challenges of Time Management and Conflicting Priorities, particularly during the month of December,” of which, 23.1% (n = 3) expressed as such. This theme emerged due to the observation that December is a period encompassing holidays when individuals often strive to fulfill their year-end objectives. Consequently, participants find themselves involved in numerous activities, as eloquently expressed by one of the caregivers:

84.6% (n = 11) of the feedback given by participants expressed the need to customize the TYB program for both stroke survivors and caregivers. This feedback led us to identify the second theme of our research, which we titled “Tailoring TYB and Enhancing Engagement for Stroke Survivors and Caregivers.” It was noted by some participants that the program seemed too generic and did not effectively resonate with certain stroke survivors:

Furthermore, caregivers also expressed their opinions regarding tailoring the program specifically for their needs:

A possible deviant case which did not fit into the theme, was expressed by a caregiver:

Stroke survivors had significantly higher depression scores (4.0 ± 8.0) than caregivers (0.0 ± 3.0) before the start of TYB (p = 0.02) (Table 3). However, there were no significant differences in depression scores between stroke survivors and caregivers after the completion of TYB (Table 4). Table 5 demonstrates a significant improvement in depression scores among stroke survivors, as the scores decreased from 4.0 ± 8.0 to 1.0 ± 6.0 after TYB (p < 0.01). Conversely, Table 6 illustrates that there were no significant changes in depression scores among caregivers before and after TYB.

Individual analysis indicated that among stroke survivors, 44.4% (n = 12) of the general population were expected to attain scores lower than the tested individual’s observed performance. Following the implementation of the program, this proportion decreased to 40.7% (n = 11). In contrast, caregivers exhibited a lower proportion of 28.6% (n = 6) both before and after the program.

Stroke survivors generally exhibited higher anxiety scores compared to caregivers, resulting in higher overall scores pre-TYB (Table 3). Although the anxiety scores did not show significant differences between the participant categories, the value of ps were very close to 0.05. Table 4 presents a significant difference in anxiety scores between stroke survivors (2.0 ± 4.5) and caregivers (1.0 ± 2.0) after the completion of TYB (p = 0.03). The anxiety scores demonstrated a decrease from pre-TYB (4.0 ± 5.5) to post-TYB (2.0 ± 4.5); however, these changes did not reach statistical significance (p = 0.19) (Table 5). Among caregivers, there were no significant changes in anxiety scores before and after TYB, as shown in Table 6.

Among stroke survivors, prior to the program’s implementation, 51.9% (n = 14) of the general population were projected to achieve scores lower than the observed performance of the tested individual. Upon introducing the program, this percentage decreased to 37.0% (n = 10). Similarly, caregivers initially exhibited a lower proportion of 38.1% (n = 8), which decreased to 28.6% (n = 6) after participating in the program.

Stroke survivors generally exhibited higher stress scores compared to caregivers, although no significant differences were found in stress scores across participant categories, the value of ps were near 0.05, as shown in Table 3. There were no significant disparities detected in stress scores between stroke survivors and caregivers as shown in Table 4. Table 5 demonstrates a significant improvement in stress scores among stroke survivors, with scores decreasing from 4.0 ± 9.5 at pre-TYB to 2.0 ± 6.5 at post-TYB (p < 0.01). Conversely, there were no significant changes in stress scores for caregivers before and after participating in the program, as illustrated in Table 6.

In examining stroke survivors, it was found that prior to the program, 44.4% (n = 12) of the general population were projected to have scores lower than the tested individual’s actual performance. However, after the program was implemented, this percentage decreased to 37.0% (n = 10). In comparison, caregivers initially had a lower proportion of 42.9% (n = 9), which further decreased to 33.3% (n = 7) following the program.

In the pre-TYB period, stroke survivors exhibited notably lower composite cognitive scores (28.2 ± 6.79) compared to caregivers (37.9 ± 9.01), with a significant difference (p < 0.001) as shown in Table 3. However, there was no significant difference in cognition between the two groups post-TYB. Table 5 reveals an enhancement in cognitive scores for stroke survivors, with an increase from 28.2 ± 6.79 to 31.5 ± 7.18 after TYB; however, the associated value of p is 0.30, indicating no statistical significance. On the other hand, Table 6 demonstrates a significant decline in cognitive scores among caregivers after TYB, as their scores decreased from 37.9 ± 9.01 to 28.1 ± 7.23 (p < 0.001).

The individual analysis revealed that, among stroke survivors, 44.4% (n = 12) of the general population were expected to have scores lower than the observed performance of the tested individual. After implementing the program, this proportion increased to 51.9% (n = 14). Conversely, for caregivers, the initial proportion showing scores lower than the observed performance was 52.4% (n = 11), which decreased to 42.3% (n = 9) following the program’s intervention. The outcomes derived from individual analyses are available for reference in the Supplementary Appendix B.

This study had several limitations that must be acknowledged. There was no measurement of the long-term effect of TYB as the post-intervention measures were taken almost immediately on completion of TYB. Hence, the maintenance of gains could not be ascertained. The DASS-21 is not a diagnostic tool for psychological disorders; results may have differed using a clinical diagnostic system and established criteria for diagnosis of depression, anxiety, and stress-related disorders. Since DASS-21 has not been validated on a sample of stroke survivors in Singapore, the applicability of the findings to the broader stroke survivor population might be limited. The small sample in this study did not allow for a control group and randomization, or for blinding of data collectors. The small sample size also reduces the power of the statistical analysis and limits generalizability to a larger population. In addition, the study did not inquire about whether the participants had received any form of social support or counseling from their respective service providers. It is important to consider these factors as these potential mediators could influence the outcomes of the study. Furthermore, this study was limited to only English-speaking participants, thus potentially different outcomes may occur for non-English speaking individuals, which is an important consideration in a multicultural society like Singapore.

In conclusion, this pilot single trial demonstrated the feasibility and potential effectiveness of the Train-Your-Brain (TYB) program, a community-based cognitive intervention for stroke survivors and caregivers. The program received positive feedback and high levels of satisfaction from participants. To better cater to the specific requirements of stroke survivors and caregivers, it is recommended that the program be customized, and that assessments be avoided in December. The study has also shed light on the importance of age-related coping strategies and cultural collectivism in influencing the outcomes of the program. Future research should consider long-term effects, larger sample sizes, and non-English-speaking populations to enhance generalizability and explore the potential benefits of cognitive-based therapy in stroke rehabilitation. Despite the limitations, these findings indicate that the TYB program holds promise in alleviating psychological distress and enhancing cognitive function among stroke survivors, underscoring the necessity for tailored interventions for caregivers.