Digital Therapeutics for Patients with Mild Cognitive Impairment: Design of a Mobile Health Program

Abstract

Background: For patients with mild cognitive impairment (MCI), a multidomain intervention that integrates physical activity, cognitive training, and dietary intervention is necessary to delay cognitive decline. The proliferation of smartphones, software development, and online service improvements now enable the delivery of online intervention programs and the concurrent collection of patient-generated data. This paper aims to report on a protocol for developing a smartphone-based multidomain intervention tailored to the level of patients with MCI. Methods: Our program targeted patients with MCI aged 60–79. The physical activity domain provides training that is appropriate for the patient’s strength and fitness levels. The cognitive training uses both verbal and nonverbal stimuli to stimulate various domains of cognitive function. It was designed to match tasks to the patient’s abilities, promoting motivational growth. Dietary interventions were designed to enable patients to plan their diet and learn to adopt appropriate eating habits. Results: We designed and organized a screen for home aerobic exercise, story quizzes, nonverbal cognitive training, and memory diary screens. Conclusions: We developed a smartphone application-based multidomain intervention program to delay cognitive decline in patients with MCI. We will verify its efficacy in future clinical trials.

1. Introduction

Alzheimer’s disease (AD) is a devastating disease characterized by progressive neuron degeneration leading to cognitive decline, which results in dependency for activities of daily living (ADL). It primarily affects the elderly population, and with the increase in average life expectancy and the growing number of elderly individuals, the prevalence of Alzheimer’s disease is expected to rise in the future [1]. Current pharmacological treatments for Alzheimer’s disease mainly provide symptomatic relief for memory impairment [2]. Although recent developments have led to the approval of disease-modifying treatments by the FDA, these medications are still limited in use and there is a lack of long-term evidence regarding their effectiveness [3]. Research has also shown that lifestyle modifications can effectively attenuate cognitive decline at the mild cognitive impairment (MCI) stage [4]. This has highlighted the growing importance of non-pharmacological treatments alongside pharmacological options. MCI is a prodromal stage of Alzheimer’s disease dementia and requires long-term management due to its progressive nature [5]. It is crucial to intervene during the mild cognitive impairment (MCI) stage to prevent progression to dementia [6,7]. However, structured offline interventions for individuals or groups may be only suitable in limited clinical environments, presenting practical challenges for the long-term maintenance of chronic conditions.

Digital health technologies have made it possible to track disease progression and implement interventions effectively for many chronic diseases. For example, digital health-care technologies have already been applied to chronic diseases such as diabetes mellitus and chronic obstructive pulmonary disease, proving to be highly beneficial [8,9]. One of the major advantages of digital therapeutics is their ability to provide immediate feedback through real-time monitoring, which can prompt behavior changes in patients.

Therefore, the best approach for treating patients with MCI is digital therapeutics that can be used in daily life. Digital therapeutics that benefit patients with MCI should demonstrate non-inferiority to existing offline programs while encouraging self-initiated cognitive improvement. Furthermore, the use of digital devices allows real-time monitoring, enabling the implementation of a level-adjustment system based on immediate usage data [10]. Through this, it becomes possible to provide content that reflects the patient’s level more accurately than offline methods.

As there are no defined methods for effectively digitizing multidomain intervention content, we aimed to identify domains where digital transformation is feasible and develop methodologies for this transition. Thus, building on insights from past offline lifestyle modification studies and the nature of the disease, we aimed to design and develop digital therapeutics that are well-suited for facilitating the formation of self-initiated cognitive improvement habits in the long-term management of MCI.

2. Materials and Methods

1. General principles

To facilitate the formation of self-initiated cognitive improvement habits, we adopted several strategies that leverage the ease of implementing digital solutions to simplify the digital transformation of multidomain interventions.

(1) Quantitative Statistics and Gamification: One important component with which patients engage is the ability to receive continuous quantitative statistics regarding their activities. We designed a system that stores patient activity data in real time and provides feedback, including a leveling-up system similar to gamification.

(2) Scoring System: To support the leveling-up system, even when the data are relatively inaccurate, we implemented a scoring system to provide evaluative feedback. This helps patients to make more positive assessments of their activity changes.

(3) Motivational Feedback for Activity: We adopted a motivational approach to maintain habit formation, which involves using easily measurable indicators such as physical activity to provide information on activities that contribute to cognitive improvement.

(4) Personalized Levels: Considering the variability among patients, we designed an initial setup to assess and set levels tailored to each patient’s cognitive function. This ensured that the difficulties appropriately matched their initial cognitive abilities. Standardized content cannot equally improve all patients’ cognitive functions; therefore, our approach aims to offer therapeutic activities closely aligned with each patient’s current cognitive abilities.

2. Overview stages

We underwent three stages in the development of the digital therapeutic application for the management of MCI. In stage 1, we conducted a targeted literature review to determine how to define the target population and identify which types of interventions are effective in attenuating cognitive impairment in patients with MCI. In stage 2, we designed interventions that are both suitable for attenuating cognitive impairment in patients with MCI and adaptable to an online format based on previous research. In stage 3, we implemented the designed interventions in a practical application.

(1) Targeted literature review

We conducted a literature review on PubMed, focusing on studies that implemented multidomain interventions or lifestyle modifications for patients with MCI. A total of 98 randomized controlled trials (RCTs) were initially retrieved. After excluding studies that did not match the target condition or intervention types, we selected 28 studies for further reference.

(2) Designing of interventions

We identified three primary domains for intervention based on standard protocols in offline intervention studies: physical activity, cognitive intervention, and dietary intervention. After discussions with technical experts, we distinguished between intervention methods that would be more effective using paper-based formats versus those better suited to digital applications. This led to the selection of intervention methods optimized for a digital environment.

(3) Development of application

Based on the designs established in stage 2, we developed a mobile application that implements the multidomain interventions. The application prototype includes a home screen and three domain-specific intervention modules: physical activity, cognitive intervention, and dietary intervention. Additionally, we developed an interface for health-care providers to issue prescriptions and adjust intervention levels, as well as a dashboard for tracking patient activities.

3. Results

1. Targeted literature review

A well-designed randomized controlled study, the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) trial, demonstrated that non-pharmacological intervention can effectively attenuate the progression of cognitive decline [11]. In the FINGER study, four areas of multidomain intervention were implemented: (1) dietary intervention—a balanced diet based on Finnish nutrition recommendations [12]; (2) physical activity—individually tailored, progressive muscle strength training and aerobic exercise programs, including exercises to maintain and improve postural balance, based on the Dose-Responses to Exercise Training (DR’s EXTRA) study protocol [13]; (3) cognitive training—targeting cognitive domains most sensitive to aging and central to daily activities, such as episodic memory, executive function, mental speed, and working memory; (4) monitoring and maintenance of metabolic and vascular factors—risk factor assessment according to the latest national evidence-based guidelines.

The effectiveness of multidomain interventions in patients with MCI has been validated in several RCTs [14]. In 2017, a worldwide FINGER network was established to verify findings across different ethnic groups [15]. Within this network, the SUPERBRAIN study, which focused on the South Korean population, also demonstrated the effectiveness of multidomain interventions [16,17]. However, the current approach relies primarily on offline multidomain interventions, with digital devices playing only a supplementary role in training [16]. Despite its advantages, offline intervention is limited by time and space, making it accessible only to a few people. Additionally, the effectiveness of training depends heavily on the skills of the trainers and the quality of the content provided, making it difficult to ensure uniformity in training quality [18]. Although physical activity and cognitive intervention programs are structured as evidence-based offline programs, there is agreement only on the cognitive domains to target, with no standardization of the content for an effective program [14,19].

2. Designing of interventions

(1) Participant definition

Our digital therapeutics target elderly patients with MCI. The inclusion criteria were as follows: (a) age 60–79 years; (b) patients with MCI meeting the diagnostic criteria for minor neurocognitive disorder as determined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) [20]; (c) absence of dementia according to physicians’ diagnosis; and (d) a Clinical Dementia Rating (CDR) [21] global score of 0.5, with a memory domain score of ≥0.5.

The exclusion criteria were as follows: (a) severe illness with an anticipated fatal outcome within three months; (b) language barrier; (c) deafness or blindness; and (d) inability to provide informed consent.

(2) Intervention

Our application consisted of three components: (1) physical training, (2) cognitive intervention, and (3) dietary intervention (Figure 1).

Before the intervention, we measured muscle strength levels using a questionnaire based on the SARC-F measure [22] and simple movements designed to assess upper body, lower body, and core muscle functions. Cognitive function was measured using the cognitive impairment screening test (CIST) [23]. Based on these results, we categorized the physical fitness and cognitive levels as high, medium, and low. Each participant started the program at a level that corresponded to their assessment results. The program lasted 12 weeks, and each participant followed a program tailored to their level throughout this period (

Table 1. Multidomain intervention.

Domain	Objective and Design
Physical activity	The physical activity program consists of aerobic exercise, resistance exercise, and stretching. Before starting the program, upper body, lower body, and core strength were measured, and a frailty survey was conducted to classify participants into upper, middle, and lower levels. The program began at the level appropriate for each participant. Aerobic exercise was provided five times a week for 30 min or more. Resistance exercise and stretching were provided three times a week for 20–30 min through video sessions.
Cognitive intervention	The cognitive intervention consists of story quizzes specialized in language memory training, cognitive training tailored to nonverbal cognitive areas, memory diaries utilizing daily episodic memory recall, and brain-challenging games as tools for constant brain activity stimulation. Memory training is provided as story quizzes adapted from fairy tales. Training for cognitive domains such as frontal/executive function and visuospatial function is provided in the form of cognitive training targeting those specific domains. For episodic memory training, participants write a diary daily. At the end of each week, they review the events of the week through quizzes based on the content of the diary and information recorded on smart devices. A total of 30 min of cognitive training is provided five times a week, adjusted to the participant’s level. The results from the week determine the level for the following week. The level indicates the current cognitive capacity, with quantitative measures for maintenance and improvement based on the degree of cognitively encouraging activities.
Dietary intervention	The dietary intervention was based on the MIND diet. Nutrition education was provided through videos, and patients selected the ingredients they consumed daily to receive a score calculated based on the MIND diet scoring system. Feedback from the scoring encouraged behavior change by motivating patients to adjust their diet independently.

).

(2-1) Physical activity

The physical activity program was designed with reference to the World Health Organization (WHO) guidelines to provide structured physical activities that can protect cognitive function in patients with mild cognitive impairment [24]. The physical activity program consisted of aerobic, muscle-strengthening, and stretching exercises (

Table 2. Components of physical activity.

	Low	Medium	High
Aerobic exercise
Minutes/session	30	45	60
Frequency, per week	5	5	5
Resistance exercise
Minutes/session	20	25	30
Frequency, per week	3	3	3
Muscle groups	Upper body, lower body, core	Upper body, lower body, core	Upper body, lower body, core
Number of sets	3	3	3
Number of repetitions per set	12	15	18
Stretching exercise
Minutes/session	5	5	5
Frequency, per week	3	3	3

).

For aerobic exercise, the program included brisk walking at a moderate-intensity level suitable for each participant, with the aim of at least 150 min per week. This was structured as brisk walking for at least 30 min a day, five times a week. Brisk walking speeds were adjusted to high, medium, and low levels, and target times were set based on each participant’s performance to ensure safe and adequate exercise. Only the time spent walking at speeds above the specified threshold for each level was counted as the recognized exercise time.

Muscle-strengthening exercises were provided three times a week, with a two-week supply of exercises given at once to allow participants to exercise more if they desired. The exercises targeted major muscle groups, with videos demonstrating the upper body, lower body, and core muscle exercises. Each session included three sets of exercises, with each set consisting of 12–18 repetitions according to the level of the participant. Stretching was included at the beginning and end of each muscle-strengthening exercise video and took five minutes to complete.

(2-2) Cognitive intervention

The cognitive intervention was initiated at the corresponding level. It consisted of 30 min of each session conducted five times a week. The focus was primarily on training memory and frontal/executive functions, with supplementary training for visuospatial and language functions (

Table 3. Components of cognitive intervention.

Cognitive Domain	Category	Structure
Episodic memory and language	Story quizzes	A story segment of about 3 min is presented. Then, a quiz related to the story content is given. After solving the quiz, the same story is presented again. Twenty minutes later, another quiz related to the story content is given, and after solving the quiz, the story is presented again with the option of listening to it as many times as desired. The next day, at the start of the program, a quiz related to the story heard the previous day is given.
Episodic memory	Memory diary	Participants write a diary every day and take a quiz on Sunday based on the content they wrote throughout the week.
Frontal executive function and attention	Nonverbal cognitive training	Classify into upper, middle, and lower levels and assign tasks according to their level.
Frontal executive function and attention	Brain-challenging games	Participants play the game to encourage cognitive function.
Visuospatial function	Nonverbal cognitive content	Classify into upper, middle, and lower levels and assign tasks according to their level.

).

Episodic memory was stimulated by story quizzes and memory diaries. The story quizzes were designed as fairy tales familiar to most participants, encouraging them to remember detailed information from the stories. This method promotes repeated listening and reviews to help participants develop a habit of learning through repetition. The memory diary involves recording daily activities such as meals, meetings, and visits. Additionally, participants were given missions to capture photos of the places they visited and the meals they ate, allowing them to leave objective photographic records in addition to relying on their memories. Once a week, the participants used the information recorded in their memory diary to take a quiz that helped them recall events from the past week. Frontal/executive and attention functions were trained using digital cognitive content. The content is flexible and can be produced in various formats. This digital content aimed to assess current cognitive abilities while the content was being consumed and to score participants. The performance and accuracy rates of the tasks completed over the week were used to determine the level for the following week. Games are also used to stimulate the frontal, executive, and attentional functions. These games are presented as additional features for patients who may feel uncomfortable with quiz-style activities. The purpose is to engage in the experience and encourage the habit of utilizing cognitively stimulating content, particularly for patients who lack cognitive training through hobbies or other activities.

(2-3) Dietary intervention

Dietary intervention was based on the Mediterranean–Dietary Approaches to Stop Hypertension Diet Intervention for Neurodegenerative Delay (MIND), which recommends a high intake of green leafy vegetables, other vegetables, berries, nuts, whole grains, fish, beans, and poultry. It also aims to limit the consumption of red meat, pastries, sweet foods, and fried foods [25]. The diet used in the program was based on the Brain Health Diet List from the South Korean Ministry of Health and Welfare [26]. Based on this, we used a scoring system proposed in previous studies to present the results as scores when a patient selected the ingredients that they had consumed [25]. This scoring-based feedback was used to encourage behavioral changes in patients, guiding them toward a healthier diet. If it is determined by scoring that a diet is sufficiently beneficial, feedback is provided to maintain the diet. Additional information will be provided in case of any specific dietary deficiencies.

3. Development of application

(1) Home Screen

The home screen is designed to allow patients to verify the representative statistics of their past activities and to provide quick access to the activities that they need to complete today. In addition, it is structured to motivate patients by reviewing activity data and providing appropriate feedback on their activities (Figure 2).

(2) Aerobic Exercise Screen

The aerobic exercise screen was designed to provide clear information on brisk-walking exercises. The target exercise amount not only displays the cumulative walking distance but also shows measurements for walking at a speed that matches the current level of the patient. By consistently walking the target distance at the appropriate speed, patients can develop the habit of brisk walking, rather than walking slowly over long distances (Figure 3).

(3) Story Quiz Screen

The story quiz screen consists of audio content that incorporates detailed information into familiar stories, allowing quiz questions. Questions on the story focused on artificially added details presented in a simple short-answer quiz format. This design helped train patients to listen attentively through simple quizzes (Figure 4).

(4) Nonverbal cognitive training screen

Mini-games were designed considering activities of daily living (ADL), acknowledging that kiosks can be challenging for patients. The games simulate real-world scenarios as closely as possible, featuring representative kiosks, such as automated teller machines and restaurant menus. By performing these tasks, patients can become more familiar with kiosks in their daily lives, thereby increasing their usage frequency. In addition, the system was designed to continuously add various mini-games that aid in cognitive function maintenance (Figure 5).

(5) Memory Diary Screen and Weekly Quiz

The memory diary was designed to allow patients to record information about their activities three times a day. To enhance convenience and encourage frequent recording, the diary uses a simple selection format, rather than requiring precise writing (Figure 6). Additionally, a weekly quiz was provided, which was randomly selected based on the stored photos and diary records. This allowed patients to check their answers based on the information they had personally entered (Figure 7).

(6) Dietary intervention

The interface was designed to allow patients to select ingredients categorized by food groups, enabling them to track the ingredients they consumed daily. After selecting the ingredients, a brain health score is displayed at the end of each meal, providing feedback on whether the meal included sufficient brain-healthy ingredients or if there are areas for improvement (Figure 8).

4. Discussion

We designed and developed a mobile application composed of physical activities, cognitive interventions, and dietary interventions aimed at attenuating the progression of cognitive decline in patients with MCI. To attenuate cognitive decline in patients with MCI, consistent nonpharmacological treatments such as lifestyle modifications are imperative, in addition to medication. However, relying solely on traditional clinical care has its limitations in consistently implementing and monitoring these nonpharmacological treatments. To address this need, we developed a mobile application with the goal of utilizing it as a form of digital therapeutics.

The level of physical activity is set to develop a habit of regularly engaging in physical activities with cognition-protective effects. According to the WHO guidelines, the best activity is to perform moderate-intensity exercise for at least 150–300 min per week [24]. However, imposing the same level of exercise on previously inactive individuals can decrease motivation and increase the risk of injury [27]. Therefore, the program starts at a level that participants can accept, and gradually increases the duration and intensity of exercise to habituate them to physical activities that have optimal cognitive protective effects. To achieve this, speeds corresponding to moderate intensity for each level were set, and only physical activities performed at or above these speeds were counted as valid physical activities. This helps the participants recognize the physical activity that contributes to cognitive protection. Additionally, the visualization of the levels aims to provide a tool for habit recognition and motivation, supporting continuous and managed physical activity. Walking was chosen as the primary form of exercise because the WHO guidelines recommend aerobic exercise of at least moderate intensity. According to the physical activity guidelines provided by the US Department of Health and Human Services, examples of moderate-intensity aerobic exercises include brisk walking, raking the yard, mopping floors, playing tennis with a partner, swimming, cycling at speeds of less than 10 miles/hour, and yoga [28]. Among these, brisk walking was considered the most sustainable option for aerobic exercise in this study because it can be measured quantitatively and does not require special equipment. Instructional videos were created to guide the muscle-strengthening exercises. These videos promote multicomponent physical activities, including muscle strengthening, for major muscle groups that are recommended for cognitive protection. Exercises were demonstrated by models of the same age group to ensure appropriate movements that did not strain the joints or muscles, providing new learning opportunities for correct muscle-strengthening exercises. The purpose was to reduce the risk of injury by maintaining proper exercise posture through guidelines and promoting efficient activities through correct physical activity habits. Each set followed the standard general exercise routines, with three sets of 12–18 repetitions per muscle-strengthening exercise [29,30]. Accordingly, exercises were designated to have 12–18 repetitions per set based on the participant’s level.

Cognitive interventions aim to improve memory and frontal/executive functions, which primarily decline in older adults [31]. Specifically, they focus on episodic memory managed by the hippocampus and frontal/executive functions managed by the dorsolateral frontal cortex [32,33,34]. To stimulate the dominant hemisphere, training involves using language and numbers to stimulate the dominant hemisphere [35]. To stimulate the nondominant hemisphere, the training involves shapes, figures, and blocks [35]. This design also provides supplementary training for visuospatial and language functions, which typically decline following memory dysfunction in AD [36]. The rationale for setting the levels in cognitive training was that the most beneficial training difficulty was slightly above the current level [37,38]. An excessively easy level fails to stimulate the brain, while a level that is too difficult can result in a loss of interest and motivation. Therefore, the levels were established to ensure the most effective implementation of cognitive training. Establishing a routine of daily short sessions, similar to habits formed through activities such as memory diaries and aerobic exercises, is crucial. Thus, cognitive training is designed to emphasize the importance of forming daily habits, even with shorter training durations.

Story quizzes were used to train episodic memory. Training participants to memorize words is a task that rarely occurs in daily life, making it generally uninteresting and resulting in extremely low adherence to treatment [39]. To consider treatment adherence, quizzes were designed in the form of fairy tales familiar to most participants, encouraging them to remember the detailed information within the stories. This method promotes repeated listening and reviewing, stimulating the habit of increasing the number of memory activities required [40]. Another aspect of episodic memory training involves the maintenance of a memory diary. This activity encouraged participants to recall daily events accurately, leading them to consciously remember their activities. The goal is not to ensure perfect accuracy, but to make it a habit of storing daily information. Additionally, once a week, the participants took a simple quiz on their long-term memory to emphasize the importance of the activity in recalling memories.

Digital cognitive content was designed to improve frontal and executive functions and attention. It was designed such that it could be produced in multiple formats. The design strategy recommends daily cognitive improvement activities of approximately 20 min and ensures that participants consistently engage with content that is slightly above their cognitive ability. This approach helps prevent periods of inactivity due to excessively difficult tasks, which could lead to frustration and abandonment, or excessively easy, which might not provide sufficient stimulation. In addition to nonverbal digital cognitive content, games were included as an additional feature for patients who may feel uncomfortable with quiz-style activities. The games used in our program were designed to improve cognitive abilities, similar to the various content developed for younger individuals. The recommended content is open-source or license-free to avoid legal issues, since there are many other programs with games aimed at improving cognitive functions. Our program aims to use games to make cognitive training more engaging and to encourage the use of a wider variety of external content. Furthermore, this approach aims to develop the habit of utilizing cognitive content, particularly for patients who lack hobbies or other activities that provide cognitive improvement.

Dietary interventions were implemented as dietary assessments. Because changing to a standard diet can be challenging to motivate because of individuals’ long-standing personal dietary patterns [41], this approach was chosen. The design strategy involved participants providing their current dietary information to receive feedback on how beneficial their diet was to their brain health. This method aims to motivate participants by providing appropriate dietary information, rather than enforcing strict dietary changes. The design focused on providing useful information, rather than obtaining precise dietary details. The participants entered their dietary information and received relevant feedback to establish a motivational system. The diet was evaluated using a brain health diet list from the Ministry of Health and Welfare and restructured based on a scoring system derived from the MIND diet [25]. Although this design can be further refined through collaboration with nutrition experts, the current focus is to deliver useful information. In addition, to address the diversity of age groups and living environments, it is essential to improve the specificity of dietary ingredient suggestions through initial information input. This will allow for more detailed and personalized feedback in the future.

In our study, we did not include vascular risk factor management because we believe that education on vascular risk factors is not sufficiently effective in a digital environment. Furthermore, we believe that interactive offline education is more suitable for managing vascular risk factors. To manage vascular risk factors with digital devices, monitoring and providing feedback would be the most effective way to induce behavioral changes. Although the use of wearable devices could be attempted in the future, it was not feasible to implement this solely with our device; therefore, it was not included.

There are some limitations in our studies. First, since the intervention was designed as a mobile application, it may present a barrier to older adults. However, we chose to develop the intervention in the form of a mobile application because mobile devices allow for the efficient collection of both active and passive data, providing valuable insights into patient behavior and intervention efficacy. Mobile devices are the most portable and effective for real-time monitoring compared to other digital devices. We believe that developing digital therapeutics using mobile devices is essential for delivering personalized and patient-centered care, which is why we decided to adhere to a mobile application format. Second, our application requires patients to enter and interact with text, which may limit accessibility for individuals who are illiterate. Although voice recognition technology is not yet sufficiently advanced to be utilized in this study, we believe that transitioning to voice-based input methods in the future could improve accessibility for older adults. Third, while we verified the data to ensure that patients were engaging in appropriate activities by using both the active data provided by the patients and the passive data obtained through the mobile device, there is always the potential for errors in the data collection process. Lastly, the specificity of the physical activity and dietary intervention programs in the application is somewhat limited. For example, in the case of strength-training exercises, changes in strength after exercise should ideally be immediately detected and reflected in the level settings, but we have not yet implemented this feature. Additionally, the dietary intervention requires more careful consideration of the ingredients that should be included in each food group, which has not been fully addressed. There is a need to further refine the content of these programs in future versions. Despite these limitations, we designed and developed a digital intervention program that aimed at attenuating the progression of cognitive decline in patients with MCI based on offline lifestyle modification and nonpharmacological treatments. The core idea behind this digital therapeutic is that motivation through evaluation and feedback will drive behavioral change. Future clinical trials are necessary to validate the effectiveness of such digital interventions, including our core concept, in attenuating the progression of cognitive decline in elderly patients with MCI.

5. Conclusions

In this study, we designed a methodology to implement a multidomain intervention using digital devices to maintain cognitive function in patients with MCI. As MCI is a chronic disease, relying solely on offline interventions makes it difficult to achieve the desired effects. Therefore, we aimed to develop a digital therapeutic program that promotes the maintenance of cognitive function through online habit formation. Personal information obtained from digital devices can be used to reflect the level of personalization, which is updated daily. Transitioning to online tools is more efficient than relying on limited offline support for multidomain interventions, as they help patients form habits. Moreover, it is necessary to conduct clinical trials to verify whether digitized multidomain interventions will show similar effectiveness to offline interventions.