Study design and procedure
The study is a single-blind, pragmatic randomized controlled trial with two parallel groups and a 2-week follow-up. Participants will participate in ten one-hour treatment sessions (see Fig. 1). During each intervention session, participants will receive 30 minutes of AACT in addition to 30 minutes of SLT (experimental group) or 60 minutes of SLT (control group). Participants in the AACT + SLT group will learn to use a paper-made (low-technology) communication board to express themselves. The communication board includes 30 pictures relating to basic needs, moods, medical conditions, and daily activities. Except for the two experimental conditions above, participants’ regular rehabilitation interventions, such as physical and occupational therapies, remain the same. To our knowledge, a parallel design has not been used before in the AAC literature. This design allows measurements pre- and post- intervention, thus providing information on improvement over time within one subject.
Setting and study population
PWA who are receiving in-patient stroke rehabilitation services, are screened for eligibility. These participants are enrolled in in-patient service in Yueyang Hospital of Integrated Traditional Chinese and Western Medicine Shanghai University of traditional Chinese medicine, Shanghai, China (YY Hospital). Table 1 lists the inclusion and exclusion criteria. We strive to include 30 participants, based on a power analysis (see Data analysis section). Before inclusion, all participants and their family members review the written information and an oral explanation from the research staff. Participants are required to review the information and sign the consent form before participating in the study. This study has been approved by the Medical Ethics Committee (MEC) of YY hospital (reference number: 2019-118, approval received in November 2019). The researchers will report serious adverse events to the MEC. The events will be managed based on institutional policies.
Table 1
Inclusion and exclusion criteria
Inclusion criteria
1. Diagnosis of aphasia, after stroke, as confirmed by a qualified speech-language therapist ascertained with the Chinese Standard Aphasia Battery (ABC) test [33]
2. Moderate to severe information reduction of Verbal production (percentile score of 50 or lower on the ABC verbal production subtest)
3. First onset of stroke, anytime post-stroke
4. Age 20–80 years
5. Mandarin Chinese as the first language. (the first) native language
6. Able to participate in intensive therapy
Exclusion criteria
1. Acute stroke with unstable vital signs
2. Premorbid speech and language disorders caused by a neurological deficit other than stroke
3. Premorbid cognitive impairment, mood disorders, mental disorders affecting communication
4. Severe non-linguistic cognitive disturbances impeding language therapy
5. Presence of serious heart, liver, kidney, infectious disease and uncontrolled medical problems
6. Concomitant neurological diseases other than stroke
|
Randomization and blinding
Participants who meet the inclusion criteria will be randomly assigned to the experimental group or the control group after the baseline assessment. To randomize participants to the experimental or control condition on the ratio of 1 to 1, we use a list of two-number codes based on a computer-generated sequence. Then these codes are ordered from the smallest to the largest and transferred to new numbers. Odd numbers are allocated to the experimental condition (AACT + SLT), and even numbers are allocated to the control condition (SLT only). A researcher, who is not involved in the assessment and training process, seals these new numbers in numbered opaque envelopes. At the beginning of each participant’s first intervention session, an SLP will open one of the envelopes, so he/she knows a participant’s assigned condition. Therefore, all participants are blinded to their treatment condition.
Intervention
In each intervention session, both groups will receive a one-hour intervention in total. The experimental group will receive AACT (0.5 hour) + SLT (0.5 hour). The control group will receive a one-hour SLT. The same SLP provides all interventions to all participants. Each intervention session is video recorded.
AACT (experimental condition) is provided by teaching participants to point to the target icons on a paper-based communication board to answer questions (e.g., “Do you have pain?”, “Where is your pain?”) in line with studies that apply AACT for PWA [5,9,34]. The communication board contains 45 icons designed for PWA to communicate with medical professionals and family members in the in-patient setting. These icons are pictures of everyday needs, clinical symptoms, types of medical care, rehabilitation services, name of people around, yes/no signs, and a rating scale, with Chinese written words on the top of each icon.
Each intervention includes three training tasks. These tasks aim to help PWA to learn how to use the communication board [25]. The first task is symbol identification. Participants are asked to identify each icon on the communication board following verbal directions from the SLP. At first, the SLP presents the verbal model plus a demonstration: the SLP saying “point to [target icon]” while physically modeling the desired response by pointing to the target icon on the communication board. After the modeling, participants are asked to follow the same directions. Next, the SLP presents a verbal cue only: the SLP saying “point to [target icon],” and then requires the participant to point to the target icon independently. The second task is to answer yes/no questions. The SLP asks a yes/no question relative to a particular icon (is this a picture of [target icon]) and then asks the participant to respond to the question by pointing to the yes or no icon on the communication board. The third task is to respond to questions with a specific icon. The SLP asks an open-ended question relative to a particular icon (e.g., “if you have a headache, which icon will be selected to tell the people who are surrounding you?”). Then the participant responds to the question by pointing to the target icon.
In SLT (control group), fifty words related to the stroke rehabilitation program's routine are selected. The fifty words include categories as vegetables and fruits, daily supplies, daily actions, body parts, the salutation of surrounding medical professionals and family members (husband/wife, son/daughter, physician, nurse, caregiver). These words are presented with a picture and a written word. The 50 items are divided into five sets (10 items/set) according to categories. Participants will practice one set in every two days.
Participants learn the semantic and phonological knowledge of 10 items and conduct auditory comprehension (selecting target item from two/four items following verbal directions from the SLP), word repetition, and naming tasks of learned items in each session. The tasks and training procedures are the same in each session. The procedures are as follows: 1) participants are taught with the semantic and phonological knowledge of the target item one by one; 2) they are asked to repeat the word with the SLP; 3) they are asked to do auditory comprehension tasks by selecting target items from two or four cards; 4) they are asked to name the target picture based on the Cueing Hierarchy [35].
Measurement instruments
Table 2 gives an overview of the measurement instruments used in the study. The primary outcome measure is the scale of the Communication of Basic Needs (CBN) on the Functional Assessment of Communication Skills for Adults (FACS) [36]. The scale and the sub-test items are translated into Chinese.
The FACS are used to measure the change of the functional communication skills for basic needs pre- and post- intervention. The FACS is a measure of functional communication for adults, which measures functional communication across four assessment domains: social communication, communication of basic needs, reading, writing, and number concepts, and daily planning. Participants use the scale to rate their performance. The response is scaled from 0 to 7, with the outcome measure from 0 to 49, and lower scores indicate more severe communication disability [37].
The secondary outcome measures are the Chinese Standard Aphasia Battery (ABC), a 10-minute clinician-patient semi-structured conversation (10-min conversation), the 10-item Hospital version of the Stroke Aphasic Depression Questionnaire (SADQH-10) [38], the Stroke-Specific Quality of Life Scale (SS-QOL) [39] and a satisfaction questionnaire. The ABC is a well-known aphasia test for evaluating PWA’s language abilities in clinical settings in China [33]. A 10-minute conversation is used to measure participants’ communication efficiency in the in-patient setting pre- and post- intervention. The participants are asked to talk about their daily lives and their medical services with the SLP. The conversation will be video recorded and analyzed by an SLP who is not involved in the intervention sessions. The SLP will evaluate the conversation based on four qualitative domains: adequacy, appropriateness, promptness, and communication sharing, using the 5-point scale of qualitative dimensions of communication in the FACS [36].
The SADQH-10 is used to evaluate depressive symptoms in PWA [38]. The SS-QOL is used to evaluate stroke-specific health-related quality of life [39]. A satisfaction questionnaire designed by the research team is used to evaluate the satisfaction of the interventions. The participants will complete the SADQH-10, the SS-QOL, and the satisfaction questionnaire. The participants’ caregivers will also complete the satisfaction questionnaire.
Table 2
Measurement instruments
1. CBN sub-test on the ASHA FACS [36,37]
2. ABC test [33]
3. 10-minute conversation
4. SADQH-10 [38]
5. SS-QOL [39]
6. Satisfaction questionnaire
|
Fig 2 shows the schedule of each outcome measure. The primary outcome measure: the CBN sub-test on the ASHA FACS, is completed baseline and post-intervention. The secondary outcome measures: the ABC, a 10-minute conversation, and the SADQH-10 are also completed baseline and post-intervention. The satisfaction questionnaire is completed after the intervention. The SS-QOL is completed after a 2-week follow-up for evaluating the quality of life after participants are discharged by the in-patient rehabilitation program. The additional file 1 is an overview of the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist.
Sample size
No large RCTs have been carried out within the field of Aphasia AACT in in-patient rehabilitation settings. Previous AAC studies used interventions not applicable to the current protocol and used small sample sizes. We are, therefore, unable to calculate prior accurate sample size. Thus, the sample size is calculated based on the primary hypothesis to detect a minimal but clinically significant difference between the experimental and the control groups on functional communication and language performance. In a previous RCT study [37], the participants were people with severe aphasia compared to people with aphasia in the overall sample. The FACS mean scores were 3.48 (SD = 0.90) and 5.78 (SD = 0.89) for the severe aphasia group and the overall sample, respectively. The differences were significant for the overall mean, t(92)=8.29, (p<0.001 ). This trial uses the two scores as the assumptions of the results of the two groups. This study uses a comparison between two means of the two-sample with the two-sided equality formulas to calculate the sample size and achieve a statistical power of 80% and a significance level of 5%. The trial must include a total of eight participants. Each group needs four participants to detect this difference. Since the power calculation only relates to the primary endpoint of pragmatic communication skills, and the target group has a high level of heterogeneity, the trial aims for a larger sample size than 8 participants. Based on the retrospective analysis of patients with aphasia in the YY Hospital, a sample of 24 participants can potentially be recruited over this period. After considering a drop out rate of 20%, the study aims to recruit 30 participants in total.
Data analysis
Outcomes of interest will be analyzed on an intention-to-treat basis. Potential differences of baseline outcomes between the groups are tested using independent t-tests for continuous variables, the Mann-Whitney U test for ordinal variables, and chi-square tests for categorical variables. Outcomes of the measures over time are compared for the experimental condition versus the control condition using repeated measurement analysis. This analysis considers the correlation of repeated measurements within the same patients and handles missing data, assuming that the data are missing randomly. The dependent variables are the outcome measures, and the independent variables are time and group assignments and the interaction between these variables. In these analyses, adjustments are made for potentially confounding variables that can be unequally distributed over the groups despite the randomization procedure. All statistical analyses are performed using the Stata statistical software. An alpha level of 0.05 is accepted as significant. The results of the statistical models are presented in the form of regression coefficients, their 95% confidence intervals, and effect sizes. The statistician who completed the data analyses is blinded to group allocation until the analysis is completed.