Evidence shows that people with dementia have a higher prevalence of sleep disturbance, cognitive decline, behavioral disturbance, and experience motor dysfunction. These symptoms are interrelated. However, few randomized controlled trial (RCT) studies implement dual-task training for mild cognitive impairment and mild dementia, especially for those who experience sleep problems and behavioral disturbances.

Therefore, this study aims to analyze the effect of dual-task training in improving global cognitive function, executive function, working memory, sleep, behavioral disturbances, balance, and flexibility among people with mild cognitive impairment and mild dementia.

The main questions it aims to answer are:

1. . Does dual-task training affect primary outcomes (global cognitive function, executive function, and working memory)
2. . Does dual-task training affect secondary outcomes (including sleep quality, behavioral disturbances, balance, and flexibility) for people with mild cognitive impairment and mild dementia?"

The length of dual-task training is 12 weeks; sessions are 3 times per week, each session lasts 45 minutes, and total sessions are 36. There are three groups of intervention (dual-task training, cognitive training, and motor training). While the control group receives the usual care. Researchers will compare the experiment and control groups to see the effect of the dual-task training.

Sample size calculation: Investigators use the G*Power sample size calculator under an F test with an analysis of variance based on some criteria, for instance: effect size 0.2 (small), power 80%, four times measurement, four groups (cognitive training, motor training, dual-task training, and control group) and α error probability of 0.05. The sample size from the calculator yields 48 in total, considering 50% attrition rate, finally 72 participants (18 participants for each group) will be recruited.

Sample recruitment: Investigator will use a CONSORT flow diagram to describe the phases of this study. A purposive sampling technique will be used in this study to specifically select the nursing home. Following this, eligibility interviews and screening test using MMSE (The Mini Mental State Examination) and The Clinical Dementia Rating (CDR) will be used to recruit participants. The participants will be recruited in this study CDR score varies from 1 to 4, and MMSE >=18. Other inclusion criteria are complain of conscious cognitive decline, participate voluntarily and sign informed consent. However, potential participants will be excluded if they diagnosed with moderate or severe dementia, those who have received cognitive training or motor training within six months, people with severe sensory function impairment (such as vision, hearing), and with limited mobility, such as those using wheelchairs or four-legged walking aids.

Randomization and Blinding: The allocation of the group to be either the experimental (dual-task training group, cognitive training group, motor training group) or control group will be conducted using permuted block randomization design with a block size of four. Randomization was carried out using a web-based randomization tool by a third party who was not involved in the intervention team. To ensure the allocation concealment, investigators use SNOSE principle (sequentially numbered, opaque, sealed envelopes), and randomization code will not be released until the participant or interventionist has started the trial. This study adopts a single-blind approach, by blinding data collector.

Baseline Data Collection Procedure: Participants who meet the inclusion criteria and agree to participate, will be requested to complete a sociodemographic characteristic questionnaire and the seven outcome measures. However, the completion of the seven outcome measures will be conducted in two days to avoid boredom among the participants. In the first period, a sociodemographic questionnaire, WCST, DS, NPI-Q will be completed. The PSQI, ISI, SFBBS, Chair sit-and-reach test (CSR) and Back scratch test completion will be conducted later in the second day at a time agreed upon by the researcher and participants.

Intervention delivery: 1). The dual-task training is a combination of cognitive tasks and motor tasks. Motor task consist of six components, including: strength, balance, agility, gait, aerobic capacity, and flexibility. Whereas Cognitive task consists of four cognitive domains: attention, executive function, learning and memory, language, and perception motor. Both component will be delivered simultaneously or at the same time, and will be delivered in a group format, consisting of 8 to 10 participants for each group. The length of therapy is 12 weeks; sessions are 3 times per week, each session lasts 45 minutes (20 minutes first: cognitive tasks and motor tasks, 5 minutes break time (at this time, participants will be given drink and snack), 20 minutes second: cognitive tasks and motor tasks), and total sessions are 36. In this study, only one interventionist (the principal investigator) from the research team will deliver the intervention sessions. The detail description of each session are below: session 1-6: Participants initially receive one of two tasks (either a cognitive or motor task) and undergo training for each individually; session 7-12: participants receive cognitive and motor task at the same time; session 13-18: repetition of sessions 7 to 12, and session 19-36 are repetition from session 1 to 18. 2). Cognitive training= in this group, participants only receive cognitive training. 3). Motor training= in this group, participants only receive motor training. The duration are same with dual-task training.

Outcomes: The primary outcomes of this study are global cognitive function, executive function, and working memory. Global cognitive is measured using MMSE that consists of five domains. MMSE 0-17 scores indicating severe cognitive impairment, 18-23 mild cognitive impairment, and 24-30 as normal (no cognitive impairment). MMSE was valid and had reliability of 0.763. Executive function in this study will measure using the Wisconsin card sort test (WCST), which has satisfactory reliability and validity for measuring executive function, possessing a split-half reliability of 0.89-0.93 for older adults and test-retest reliability of 0.76. Working memory is measured using a Digit span (DS) test that comprises both digit forward (16 points) and digit backward (14 points). The secondary outcomes of this study are sleep quality, behavioral disturbance, balance, and flexibility. Sleep quality is measured using the Pittsburgh Sleep Quality Index (PSQI) and Insomnia Severity Index (ISI). Behavioral disturbance will be assessed with the Neuropsychiatric Inventory-Questionnaire (NPI-Q). Balance is measured with the Short Form Berg Balance Scale (SFBBS), and flexibilty using flexibility using the Chair sit-and-reach test (CSR) and Back scratch test.

Post-test and Follow-up Data Collection Procedure Participants will be requested to complete questionnares consist of: MMSE, WCST, DS, PSQI, ISI, NPI-Q, SFBBS, Chair sit-and-reach test (CSR) and Back scratch test, in the middle session of the intervention (week 6, after session 18), after intervention completed (week 12, 1 month follow-up time (week 16, measured from baseline)

Statistical analysis: The collected data will be analyzed using IBM SPSS for Windows version 23 and assumed a statistical significance level of p ≤ 0.05. Data consistency checks will be conducted to verify the reliability of the data collected. Descriptive statistics (frequency, percentage, mean, and standard deviation) will be used to summarize the characteristics of the participants and the outcomes. To compare the baseline group differences, investigators will use Chi-square or Fisher exact tests for categorical data, and independent t-tests for continuous data with normal distribution. The generalized estimating equation (GEE) test is used for analyze main outcome with follow the intention-to-treat (ITT) analysis principle, and this test can handle missing data, so no need imputation for missing data.

Interim Analysis and Stopping Guideline: During the study and subsequent follow-up data collection, investigators do not expect any issues that would be harmful to the participants. Investigators have no established termination criteria, and investigators do not plan to perform interim analyses before the follow-up data collection is completed.

Inclusion Criteria:

• Older adult >= 60 years and able to communicate
• CDR score range from 1 to 4
• MMSE score >= 18
• Complain of conscious cognitive decline.
• Participate voluntarily and sign informed consent.

Exclusion Criteria:

• Diagnosed with moderate or severe dementia.
• Received cognitive training or motor training within six months.
• Severe sensory function impairment (such as vision, and hearing).
• Have limited mobility, such as those using wheelchairs or four-legged walking aids.

Particular data will be shared: Individual participant data that underlie the results reported in this article, after identification (text, tables, figures, and appendices).

Beginning 6 months and ending 36 months following article publication.

Investigators whose proposed use of the data has been approved by an independent review committee ("learned intermediary") identified for this purpose.

Outcomes Assessor will be blind, so they didnt know which is intervention group or control group