Feasibility and effectiveness of a handbook-based multidomain lifestyle intervention for older adults with mild cognitive impairment: A quasi-experimental design with propensity score matching

Abstract

Background

The increasing prevalence of dementia highlights the urgent need for the social implementation of nonpharmacological interventions for its prevention. To address professional staff shortage challenges, we developed the MCI Handbook, a structured resource designed to guide non-professionals in promoting dementia prevention.

Objective

We aimed to evaluate the feasibility and effectiveness of MCI Handbook-based multidomain lifestyle interventions provided by nonprofessionals.

Methods

A 12-month intervention was implemented for older adults with mild cognitive impairment (MCI) recruited from two facilities in Japan. The feasibility outcomes included class participation rates and satisfaction with the intervention measured using the Japanese version of the 8-item client satisfaction questionnaire (CSQ-8J). Effectiveness was assessed based on changes in the Japanese version of the Montreal Cognitive Assessment (MoCA-J) scores.

Results

In total, 37 participants (16 males and 21 females; mean age, 78.6 years) were included. The mean class participation rate was 88.3 ± 17.7%, with high satisfaction reported by the participants (average CSQ-8J score, 25.6 ± 3.4). The MoCA-J score improved significantly from 21.9 ± 2.9 at baseline to 23.3 ± 3.8 at 12 months (p = 0.007), highlighting the clinical relevance of this intervention. Compared with the external control group, the intervention group showed significantly greater cognitive improvements (p = 0.017).

Conclusions

These findings demonstrate the feasibility and effectiveness of a 12-month multidomain lifestyle intervention conducted by nonprofessionals using the MCI Handbook. This program offers a promising, resource-efficient approach to dementia prevention. Future studies are needed to refine the delivery strategies and explore scalable methods for the broader implementation of multidomain interventions for dementia prevention.

Keywords

Alzheimer's disease cognitive function dementia feasibility mild cognitive impairment non-pharmacological intervention non-professional intervention

Introduction

Dementia is a serious public health challenge in an aging society because it interferes with a person's independence and increases the cost of both formal and informal care.¹ Since no curative treatment for dementia has been established, preventive measures and risk reduction before dementia onset are particularly important. The development of anti-amyloid agents has provided a new option for pharmacological interventions to treat Alzheimer's disease.² However, their public dissemination has not yet been achieved due to the limited number of target patients receiving the medication and the need for rigorous monitoring during treatment.³ In addition, as the number of people with dementia or those at risk of dementia is increasing—especially in low- and middle-income countries—the development and implementation of economically effective non-pharmacological interventions comprise an urgent issue.^4,5

As a non-pharmacological approach to dementia prevention, a multidomain intervention that comprehensively addresses various risk factors associated with the development of dementia (i.e., physical inactivity, social isolation, hypertension, and a depressive mood) is a representative option.^6,7 In particular, the efficacy, safety, and cost-effectiveness of the FINGER model, which consists of exercise, nutrition, cognitive training, social participation, and vascular risk management as central components, have been reported.^8,9 In our randomized controlled trial (RCT), the J-MINT study, we reported the efficacy of a multidomain intervention for older Japanese adults with mild cognitive impairment (MCI) who maintained good adherence.¹⁰

Although the efficacy of the intervention was demonstrated, several factors need to be improved regarding the program delivery methods and costs to achieve the public dissemination and social implementation of multidomain interventions. As one of the most important barriers, interventions should be repeated over a prolonged period for a sustainable effect,⁷ requiring the continuous involvement of medical staff with expertise in dementia.

In recent years, the effectiveness of peer education programs based on the peer training model has been demonstrated in several areas, such as smoking cessation, which has attracted attention to educational interventions administered by non-professionals.¹¹ Peer education programs can potentially become commonplace in dementia prevention similarly. Nevertheless, this approach is not widespread owing to requirements pertaining to the selection, training, and supervision of peer educators, the type of intervention, and the relationship between peer educators and the target education group.¹² To promote the training of peer educators for dementia prevention, we developed an MCI Handbook to help non-professionals quickly grasp the key points needed to prevent dementia.¹³ This study tested the feasibility and effectiveness of implementing nonprofessional-led classes using this handbook.

Methods

Design

This was a 12-month, multi-center, single-arm intervention trial conducted at two sites in Japan: the National Center for Geriatrics and Gerontology (NCGG) and the Tokyo Metropolitan Institute for Geriatrics and Gerontology (TMIG). All procedures involved in this study were reviewed and approved by the ethics committee of NCGG (approval no. 1603). The study was registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN) before participant recruitment (registry no. UMIN000048338). In addition, this study was conducted in accordance with the updated CONSORT statement for randomized trials of nonpharmacological treatments and guidelines for reporting non-randomized studies.^14,15

Participants

Community-dwelling older adults aged 65–85 years with MCI were included in this study. We used the Japanese version of the Montreal Cognitive Assessment (MoCA-J) to screen for MCI and defined participants as those who scored less than 26, in accordance with literature validated in a Japanese setting.¹⁶ Participants with disabilities that could have prevented the intervention, those diagnosed with dementia, or those with moderate-to-severe cognitive decline (Mini-Mental State Examination score less than 21 points) were excluded.¹⁷ The sample size was set based on a previous study showing that a 6-month exercise intervention improved the MoCA-J by 2.2 ± 2.9 points (effect size d = 0.75).¹⁸ Assuming that a similar level of efficacy was expected in the present study, we determined that 26 cases would be required under the conditions of a two-tailed significance level of 0.05 and a power of 0.95. Therefore, we planned to recruit more than 33 participants from the two sites, considering a 20% dropout rate at the final follow-up at each study site.

To recruit participants, we sent mail to those who had visited the memory clinic (at the NCGG site) or those who had participated in a community health class (at the TMIG site) and utilized the eligibility and exclusion criteria. A written informed consent form was provided to the participants who met the criteria and agreed to participate in the study. The purpose, properties, and potential risks of the study were thoroughly explained, and written consent to participate was obtained before the intervention was initiated.

Intervention

All participants received multidomain interventions according to the MCI Handbook developed by the research team.¹³ The MCI Handbook was developed to integrate information about lifestyle modifications and psychological support strategies for individuals with MCI. This handbook comprises nine domains, encompassing 38 selected patient questions (PQs): MCI, lifestyle, lifestyle-related diseases, exercise, nutrition, social participation, cognitive training, psychological care, and family support (Stable1). The accessibility and utility of the handbook were evaluated and validated based on clear communication index scores. The MCI Handbook, in PDF format (in Japanese), is available on the following web page: https://www.ncgg.go.jp/ncgg-overview/pamphlet/documents/mcihandbook-v2.pdf.

The research secretariat considered research staff members who met the following criteria to be non-professionals and assigned them the instructor role: 1) They had not previously attended courses offered by societies or research groups related to dementia and its prevention; 2) They had never hosted or participated as instructors in classes for people with MCI or dementia. One-day of hands-on training covering the study purpose, detailed intervention content, and relationship-building with participants was provided to the instructors before the start of the study to standardize the intervention using the MCI Handbook. Learning methods using the MCI Handbook were also introduced on the same day, and instructors continued to use the handbook throughout the intervention period. Instructors were offered a salary in compliance with the institute's regulations.

During the 12-month intervention period, the participants engaged in group classes organized by the instructors and self-monitored their daily lives. The research secretariat staff, with medical or co-medical licenses, participated in all group classes to motivate instructors and ensure participants’ safety. They qualitatively assessed classroom fidelity to optimize the instructors’ interventions.

Group classes. Group classes were conducted once every 2 weeks during the intervention period, with a total of 24 sessions. Each session lasted 90 min and consisted of 60–75 min of group exercise and 15–30 min of group work. Each classroom included fewer than 20 participants to ensure class quality. The group class program is presented in Supplemental Table 1.

Group exercises included stretching, aerobic exercise, strength training, balance training, and dual-task exercises (combining exercise and cognitive tasks). The exercise program was designed to increase the intensity gradually, and the instructor adjusted for each session to ensure that the participants’ subjective exercise burden was in the target range (3–6 on the 10-point scale).¹⁹ In the group sessions, instructors gave 15–30-min presentations about nutrition, exercise, cognitive training, lifestyle, disease, and psychological support for dementia prevention following the MCI Handbook. The guidelines included encouraging physical activity, including exercise at least three times per week, diverse food intake, cognitive training, social activities in daily life, and blood pressure and weight control.^10,13 A cognitive-behavioral therapy session was also conducted during 8 of the 24 classroom sessions to promote friendly relationships between instructors and participants and to consolidate information learned from the presentations. All participants briefly shared their thoughts on topics such as goal setting and addressing age-related issues (e.g., forgetfulness, decreased motivation, and social isolation) and then engaged in discussions with each other or with the instructor. The cognitive-behavioral therapy guidebook prepared by the research secretariat (Supplementary Psychological Exercises for Workouts: A Guidebook) is available here: https://sites.google.com/view/tmig-psycho-research/%E3%83%9B%E3%83%BC%E3%83%A0. The group work was followed by a few minutes of group discussion. During the group discussions, the participants discussed or reflected on what they did not understand, their life situations, and their opinions with each other and the instructor.

Self-monitoring. In addition to the group classes, we recommended self-guided multidomain lifestyle interventions such as two to three times per week of home exercise, a well-balanced diet, social participation, and cognitive training. We also distributed monitoring notebooks and pedometers to all participants to motivate them and improve their adherence to the self-guided interventions. The participants recorded their daily subjective well-being, number of steps taken, exercise activities, and diet in a monitoring notebook. The instructor reviewed the monitoring notebooks in each group classroom session, provided feedback to the participants, and encouraged healthy daily living.

Outcomes

We obtained the feasibility and effectiveness outcomes during the intervention trials. Moreover, the participants’ background characteristics, such as their age, sex, marital status, living situation, alcohol consumption, smoking status, and use of medication, were investigated using a questionnaire before the intervention. We also examined registration in the Biobank for NCGG-site participants and investigated apolipoprotein E (APOE) genotypes with the participants’ consent.

Feasibility outcome. The feasibility of the intervention was measured based on satisfaction with the group class, the class participation rate, and adverse events. Satisfaction with the group class was measured at the end of the intervention using the Japanese version of the 8-item client satisfaction Questionnaire (CSQ-8J).²⁰ The CSQ-8J consists of eight questions pertaining to the following: 1) quality of service; 2) type of service desired; 3) met needs; 4) whether they would recommend to a friend; 5) amount of help; 6) improvement in self-efficacy; 7) overall satisfaction; and 8) whether they would return. All items were rated on a four-point scale: 1, poor; 2, fair; 3, good; and 4, excellent. Total scores ranged from 8 to 32, with higher CSQ-8J scores indicating greater satisfaction with the intervention or treatment. Class participation rates and adverse events were recorded as often as appropriate throughout the intervention. Adverse events were defined as grade 3 or higher according to the Common Terminology Criteria for Adverse Events, version 5.0.²¹

Effectiveness outcome. Trained research staff conducted effectiveness outcome measurement at baseline and at the end of the intervention (12 months) in accordance with a measurement procedure guide predetermined by the research office. The primary effectiveness outcome was 12 months of changes in global cognitive function, as measured using the MoCA-J. Four physical examinations and nine questionnaire-based metrics were used as the secondary outcomes.

Independent research staff or researchers performed MoCA-J testing and scoring separately for blinding. MoCA-J scores range from 0 to 30, with higher scores indicating better cognitive performance.²² BMI was calculated from direct measurements of height and weight. Systolic and diastolic blood pressures were measured in a resting state using an automatic sphygmomanometer. Grip strength was measured once for each arm in the standing position using a Smedley-type digital hand dynamometer (TKK 5401 Grip-D; Smedley, Takei, Tokyo, Japan), and the maximum value of the two measurements was used. Gait speed was measured twice at the participants’ usual speed with a 2.4 m walking path that included a 1 m acceleration and deceleration zone each, and the average of the two measurements was used.¹⁰

The Barthel index, Lawton index, fall risk index, dietary diversity score, mini nutritional assessment short-form, 15-item geriatric depression scale (GDS-15), EuroQol 5-dimensions, 6-item Lubben Social Network Scale, and Japanese version of the Pittsburgh Sleep Quality Index were obtained using questionnaires. The Barthel Index reflects the ability to perform basic activities of daily living. Barthel Index scores range from 0 to 100 in 5-point increments, with higher scores indicating greater basic activities of daily living ability.²³ The Lawton Index measures the instrumental activities of daily living performance and consists of eight items. Female participants completed all items, whereas male participants completed five items, excluding laundry, food preparation, and housekeeping.²⁴ We used sex-standardized scores in the analysis according to the following formula:

S t a n d a r d i z e d L a w t o n I n d e x = \frac{T o t a l s c o r e o f e a c h p a r t i c i p a n t - M e a n v a l u e o f b a s e l i n e s c o r e s b y s e x}{S t a n d a r d d e v i a t i o n o f b a s e l i n e s c o r e s b y s e x}

The fall risk index examines a participant's overall risk of falling; participants answer “yes” or “no” to 21 items, with higher scores indicating a higher risk of falling.²⁵ The dietary diversity score was calculated as the sum of the 13 food items eaten almost every day (1 point), once every 2 days (0.5 points), once or twice per week (0.25 points), and almost never (0 points). Higher dietary diversity scores indicated more diverse food intake.¹⁰ The Mini Nutritional Assessment-Short Form was used to measure the nutritional status. It consists of six items with scores ranging from 0 to 2 or 0 to 3, resulting in scores ranging from 0 to 14. Higher scores indicate a better nutritional status.²⁶ GDS-15 scores range from 0 to 15, with higher scores indicating a severe depressive mood.²⁷ To measure the EuroQol 5-dimensions score, the participants responded to each of the five health-related quality of life (QoL) questions on a 3-point scale. The responses were scored as continuous variables based on a conversion table from a previous study. A higher score indicates a higher QoL, and a score of 1 indicates that the participant's QoL is completely preserved.²⁸ The 6-item Lubben Social Network Scale examines the extent of social networking by asking how many families and/or friends can build relationships in daily life. The scores range from 0 to 30, with a higher score representing a better social network.²⁹ Japanese version of the Pittsburgh Sleep Quality Index scores range from 0 to 21, with higher scores indicating worse sleep habits.³⁰

External control data

We curated follow-up data from an existing clinical trial (the J-MINT study) to select external controls. In this 18-month multi-center RCT, older adults with MCI—identified by age- and education-adjusted cognitive decline of 1.0 standard deviation or more below the reference threshold from hospitals, memory clinics, and/or community-based cohorts—were subjected to a multidomain intervention or a health education program (control intervention). Details of the J-MINT study can be found elsewhere.¹⁰ The J-MINT follow-up study was initiated immediately after the J-MINT study was completed, and participants were subjected to annual cognitive function testing. External controls were defined as those allocated to the control group in the J-MINT study who 1) participated in the follow-up, 2) had at least two MoCA-J scores at the annual follow-up assessment, and 3) had not received any intervention during the follow-up period. In this study, available data from 2022–2023 were used as external controls; the MoCA-J measurement methodology was standardized using the same manual for this study and the J-MINT follow-up study.

Statistical analyses

Participants who completed at least one intervention were defined as the full analysis set (FAS), the primary population analyzed in this study. We calculated summary statistics at baseline and 12 months for all feasibility and effectiveness outcomes. For the effectiveness outcomes, paired t-tests were performed for the pre–post comparisons. Subgroup analyses for the change in MoCA-J scores were conducted in the same manner for the following subgroups: by age at enrollment (65–74, 75 years old or older); by sex; by years of education (≤12 or ≥13 years); by intervention site; and by APOE status (APOE ε4 allele carrier or not, for participants from the NCGG).

In addition, we compared the results of the current trial with data from an external control group. Propensity score matching was performed to balance current and external control datasets. Propensity scores were estimated using a logistic regression model that included age, sex, years of education, baseline MoCA-J scores, and baseline gait speed as covariates. A 1:1 nearest-neighbor matching method without replacement was applied, and the caliper was set to 0.25. After propensity score matching based on the two datasets, we compared the 12-month changes in all outcomes using a two-sample t-test. Subgroup analyses for changes in the MoCA-J scores were conducted using the same techniques; age at enrollment, sex, and years of education were used as subgrouping factors. In addition, as a sensitivity analysis, the inverse probability weighting (IPTW) method based on the propensity score was used to estimate the average treatment effect for the treated group. The weighted populations were compared using linear or logistic regression analyses. For the IPTW analysis, standard errors were estimated using a robust sandwich variance estimator.

We did not perform a multiplicity adjustment for all analyses. Statistical analyses were performed using SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA) according to the statistical analysis plan separately defined by the statistical analyst (FK). Statistical significance was set at a two-tailed value of p < 0.05.

Results

Participants

The 64 participants (23 at the NCGG sites and 41 at the TMIG sites) who responded to the invitation were evaluated for eligibility. Through an eligibility assessment, 26 individuals (15 with MoCA-J scores > 26, two requiring long-term care, and nine randomly with a lottery) were excluded. Therefore, 38 patients (20 from NCGG sites and 18 from TMIG sites) were enrolled in the study (Figure 1). To accommodate a classroom limit of fewer than 20 participants, two and one classroom sessions were held at the NCGG and TMIG sites, respectively.

Figure 1.

Study flow diagram.

One participant dropped out before the study began owing to back pain, and three dropped out during the intervention period because of a lack of motivation, aggravation of physical symptoms, and femoral fracture occurring in situations unrelated to the classroom session. Thus, effectiveness outcomes were analyzed for 37 participants identified as having the FAS, and feasibility outcomes were analyzed for 34 participants who completed the final evaluation. Table 1 shows the baseline characteristics of the FAS;16 participants (43.2%) were male, and the mean age was 78.6 ± 4.3 years.

Table 1.

Basic demographics of all participants in this study.

	Before propensity score matching				After propensity score matching
	Current trial (N = 37)	J-MINT (N = 49)	SMD	p	Current trial (N = 19)	J-MINT (N = 19)	SMD	p
Age (y)	78.6 ± 4.3	74.2 ± 4.4	1.01	<0.001	76.9 ± 4.8	76.0 ± 4.7	0.17	0.613
Sex, Male (n)	16 (43.2)	31 (63.3)	0.41	0.082	10 (52.6%)	11 (57.9%)	0.11	1.000
Education (y)	12.6 ± 2.2	12.6 ± 2.4	0.02	0.922	12.5 ± 2.5	13.1 ± 2.7	0.20	0.542
Baseline MoCA-J score	21.9 ± 2.9	24.3 ± 3.8	0.71	0.002	22.5 ± 3.0	22.4 ± 4.4	0.03	0.932
Baseline gait speed	1.3 ± 0.3	1.2 ± 0.2	0.38	0.078	1.3 ± 0.3	1.3 ± 0.2	0.04	0.893
Marital status (n)			0.08	0.901			0.37	0.714
Married	26 (72.2)	37 (75.5)			15 (78.9)	13 (68.4)
Divorce or widowed	9 (25.0)	11 (22.5)			4 (21.1)	5 (26.3)
Never married	1 (2.8)	1 (2.0)			0 (0.0)	1 (5.3)
Living alone (n)	9 (25.0)	8 (16.3)	0.22	0.413	4 (21.1)	4 (21.1)	<0.01	1.000
Alcohol consumption (n)			0.18	0.762			0.84	1.000
Never	21 (58.3)	27 (55.1)			11 (57.9)	10 (52.6)
Sometimes	8 (22.2)	9 (18.4)			4 (21.1)	5 (26.3)
Daily	7 (19.4)	13 (26.5)			4 (21.1)	4 (21.1)
Smoking (n)			0.68	0.012			0.84	0.042
Current smoker	0 (0.0)	4 (8.1)			0 (0.0)	0 (0.0)
Past smoker	5 (13.9)	16 (32.7)			1 (5.3)	7 (36.8)
Never smoker	31 (86.1)	29 (59.2)			18 (94.7)	12 (63.2)
Medication (n)	3.3 ± 2.7	3.1 ± 2.6	0.07	0.744	3.3 ± 2.8	3.4 ± 3.1	0.02	0.957

The basic demographics of participants in the current trial and the J-MINT study (external controls) are shown. All variables before and after propensity score matching are presented. Continuous and categorical variables are presented as mean ± standard deviation and frequencies (percentages). p values were calculated using a two-sample t-test or Fisher's exact test. SMD: standardized mean differences; MoCA-J: the Japanese version of Montreal cognitive assessment.

Feasibility of the handbook-based intervention

The mean classroom participation rate was 88.3 ± 17.7% for all participants (92.6 ± 10.7% at NCGG sites and 84.0 ± 22.1% at TMIG sites). Overall, the CSQ-8J score was 25.6 ± 3.4 points on average. The mean values for all eight CSQ-8J sub-items ranged from good to excellent; the lowest score was 3.06 for meeting needs and returning, whereas the highest score was 3.59 for improvement in self-efficacy. A similar trend was observed based on each study site (Figure 2A, B).

Figure 2.

Average scores of the eight subitems of the client satisfaction questionnaire. Error bars indicate the standard deviation. Panel A shows the results for all participants, and panel B presents the results for each study site.

There were five adverse events during the intervention period, as follows: lumbar spine compression fracture, laparoscopic surgery for a left inguinal hernia, gamma knife for an auditory nerve tumor, left femur fracture, and hospitalization for spinal canal stenosis; however, for all of those, no causal relationship with the study intervention was found.

Effectiveness of the intervention

Pre–post comparison for the current single-arm population. The MoCA-J score improved from 21.9 ± 2.9 at baseline to 23.3 ± 3.8 at 12 months (p = 0.007, effect size = 0.41). No secondary outcomes showed statistically significant changes, although systolic blood pressure, dietary diversity, and GDS-15 scores were improved (p < 0.1) (Table 2). The subgroup analysis identified better responses to the intervention, in terms of cognitive functions among participants aged ≥75 years (N = 32, from 21.6 ± 3.0 to 23.0 ± 3.8, p = 0.014) and among females (N = 21, from 22.5 ± 2.7 to 24.5 ± 3.0, p = 0.015). We could not detect the effect of APOE status because only one participant carried the APOE ε4 allele (Supplemental Table 2).

Table 2.

Changes in all efficacy outcomes of the full analysis set.

	Baseline (N)		12month (N)		Change	p
Japanese version Montreal cognitive assessment	37	21.9 ± 2.9	34	23.3 ± 3.8	1.4 ± 2.9	0 . 007
Barthel index	36	99.9 ± 0.8	34	98.7 ± 6.1	−1.3 ± 6.1	0.212
Sex-standardized Lawton index	36	0.1 ± 0.8	34	−0.1 ± 1.7	−0.2 ± 1.6	0.544
Body mass index (kg/m²)	37	21.7 ± 3.2	34	22.3 ± 3.0	0.2 ± 0.8	0.179
Systolic blood pressure (mmHg)	37	146.2 ± 22.9	34	137.9 ± 17.0	−7.3 ± 21.6	0.056
Diastolic blood pressure (mmHg)	37	83.6 ± 15.7	34	80.8 ± 10.5	−2.2 ± 13.0	0.337
Grip strength (kg)	35	25.3 ± 7.4	34	26.0 ± 5.7	0.3 ± 2.8	0.565
Gait speed (m/s)	37	1.3 ± 0.3	34	1.3 ± 0.2	−0.0 ± 0.3	0.769
Fall risk index	36	7.1 ± 3.1	34	6.4 ± 3.3	−0.6 ± 2.5	0.161
Dietary diversity score	36	10.1 ± 2.1	34	10.6 ± 2.3	0.6 ± 1.9	0.095
Mini nutritional assessment short-form	36	11.4 ± 2.2	34	11.5 ± 1.9	0.0 ± 1.6	0.913
15-item geriatric depression scale	36	4.1 ± 3.3	34	3.1 ± 2.8	−0.8 ± 2.6	0.075
EuroQol 5-dimensions	36	0.9 ± 0.1	34	0.9 ± 0.1	0.0 ± 0.1	0.362
6-item Lubben social network scale	36	14.2 ± 6.2	34	14.8 ± 5.7	0.5 ± 3.6	0.454
Japanese version Pittsburgh sleep quality index	36	4.8 ± 2.0	34	4.6 ± 2.5	−0.2 ± 1.7	0.556

All variables are presented as mean ± standard deviation. p values by paired t-test are shown, and less than 0.05 are bolded.

Comparison of external propensity score-matched population. The 49 eligible participants from the J-MINT study, as an external control group, were matched based on propensity scores, and 19 participants in each intervention and control group were matched. After matching, all background factors, including age, sex, and years of education of the matched cohort, showed no statistically significant differences; therefore, it was considered that the matching was well-balanced. MoCA-J scores at 12 months changed from 22.5 ± 3.0 to 23.1 ± 4.5 for the intervention group and from 22.4 ± 4.4 to 20.9 ± 5.0 for the control group, representing a statistically significant difference (p = 0.017). In the subgroup analysis of age, sex, and years of education, no statistically significant differences were found for the MoCA-J score or any other secondary outcomes (Table 3 and Supplemental Table 3). Sensitivity analysis using the IPTW method based on the propensity score also showed a significant improvement in the MoCA-J scores (Supplemental Table 4).

Table 3.

Overall and subgroup changes in MoCA-J score in the current trial and external control after propensity score matching.

	Current trial					External control					p
	N	baseline	N	12month	Change	N	Baseline	N	12month	Change
Overall	19	22.5 ± 3.0	18	23.1 ± 4.5	0.7 ± 3.1	19	22.4 ± 4.4	19	20.9 ± 5.0	−1.5 ± 2.3	0 . 017
Age group
<75 years old	5	23.8 ± 1.1	5	25.2 ± 3.1	1.4 ± 2.9	9	22.4 ± 3.3	7	20.9 ± 4.2	−1.6 ± 2.1	0.067
75 years old or older	14	22.1 ± 3.3	13	22.3 ± 4.8	0.5 ± 3.2	10	22.4 ± 5.5	12	20.9 ± 5.7	−1.5 ± 2.5	0.107
Sex group
Male	10	21.2 ± 3.6	10	21.6 ± 4.6	0.4 ± 2.7	11	22.4 ± 3.9	11	20.8 ± 4.8	−1.5 ± 2.8	0.124
Female	9	24.0 ± 1.1	8	25.0 ± 3.9	1.1 ± 3.6	8	22.5 ± 5.3	8	21.0 ± 5.7	−1.5 ± 1.7	0.086
Education group
<13 years	13	21.6 ± 3.2	12	22.2 ± 5.2	0.8 ± 3.7	10	21.6 ± 4.5	10	20.3 ± 4.7	−1.3 ± 1.8	0.112
13 years or more	6	24.5 ± 0.5	6	25.0 ± 1.8	0.5 ± 1.4	9	23.3 ± 4.5	9	21.6 ± 5.6	−1.8 ± 2.9	0.103

All variables are presented as mean ± standard deviation. p values calculated by a two-sample t-test are also presented, and less than 0.05 are bolded.

Discussion

This study investigated the feasibility and effectiveness of a 12-month MCI Handbook-based multidomain intervention led by non-experts. The results of this study on older adults with MCI showed that the participants maintained high adherence and satisfaction and significantly improved their global cognition. Adherence in this study was comparable to or higher than that reported in previous studies conducted using group-class-type interventions for dementia prevention.¹⁰ Enhancing adherence is important because it mediates the intervention effects.^10,31 Several factors influenced the older adults’ adherence to group classes, including participants’ individual and instructor factors. Specifically, the participants’ housing, education, mental well-being, group cohesion, attitude, and instructor personality affect their short (3-month)- and medium (6-month)-term adherence.³² Moreover, previous analyses of multidomain interventions for dementia prevention revealed that the intervention type, intensity, and delivery method also influence participant adherence.³³ To ensure that older adults adhere to the intervention, face-to-face contact between participants and instructors should be maintained, and instructors should ensure that the tools used in the intervention are appropriate for older adults. In particular, cognitive decline is a significant inhibitor of the adaptation to digital tools in older people,³⁴ which may emphasize the usefulness of paper-based tools targeting MCI in classrooms. Additional motivational interventions could also be considered for those at the highest risk of poor adherence.³³

In the current study, it was possible to fine-tune the intensity and difficulty of the exercises and group work owing to the small group sizes used in the program. Moreover, the inclusion of cognitive behavioral therapy in the program may have strengthened participant–participant and participant–instructor cohesion. Furthermore, the use of paper-based booklets (the MCI Handbook and monitoring notebooks) rather than digital devices may have made it easier for older adults to become familiar with the program. Conversely, during the phase of social implementation, it is important to flexibly consider the method of program delivery—particularly the potential of digitally-assisted interventions—by considering the diversity of individuals with MCI, instructors, and the various situations in which classes are offered.

This MCI Handbook-based multidomain intervention improved the MoCA-J score by 1.4 points in older people with MCI, whereas the external control group showed a 1.5-point decline over 12 months. This 1.4-point improvement was greater than 1.0, the minimal clinically important difference calculated using the distribution method with standard deviations in previous studies, highlighting the clinical significance of this intervention.³⁵ This effect was comparable to that reported in a previous study in which a 6-month exercise intervention for older adults—including 34.8% healthy subjects—improved MoCA scores by 2.0 points.¹⁸ A novel finding of this study is that MCI Handbook-based multidomain interventions can drive improvements in cognitive function comparable to those in previous studies, even with non-professional instructors. In recent years, the trend toward personalized medicine has been gaining momentum even in the field of dementia prevention. The results of this study suggest the potential for implementing both expert-led personalized programs and non-expert-led peer education programs in the future.

The mechanism by which cognitive functions improve with multidomain interventions is unclear; however, it has been hypothesized that changes in physical and psychological factors may be associated with such cognitive changes.¹⁰ Previous studies have shown that frailty status, lower extremity function, and dietary habits improved significantly, along with improvements in cognitive function.^8,10 In this study, no statistically significant effects were observed on secondary outcomes such as systolic blood pressure, dietary diversity, or GDS-15, possibly due to limited statistical power resulting from the small sample size; however, an improvement trend was observed. Thus, the handbook-based intervention in this study may have positively affected these secondary outcomes. The combination of these minor improvements may be associated with improvements in cognitive function; however, cognitive function is a complex outcome influenced by multiple factors, and this relatively small study could not detect the effects on these outcomes. Therefore, there is a need to develop appropriate surrogate factors that are sensitive to and predictive of changes in cognitive function caused by multifactorial interventions.

The non-professional instructors who led the classrooms in this study were primarily responsible for providing general lifestyle guidance and motivation to maintain a healthy lifestyle. Whereas they cannot replace the role of primary care physicians and others, such as medical care for participants with hypertension and diabetes, instructors can adequately offer general guidance and companionship support, indicating the possibility of substantially reducing dementia-related costs. The increase in formal/informal care costs associated with dementia is a global problem, with estimates of $196 billion (95% confidence interval: $179 billion to $213 billion) and $450 billion ($424 billion to $478 billion) for the United States in 2020.³⁶ As the cost-effectiveness of multidomain interventions for dementia prevention has been demonstrated,⁹ the broad dissemination of this concept would reduce dementia-related costs. This study examined the feasibility and effectiveness of a handbook-based intervention designed to accelerate the dissemination of multidomain interventions for dementia prevention. A future trial or real-world data evaluation to assess the cost-effectiveness of a multidomain dementia prevention intervention led by non-experts is awaited.

This study had some limitations. First, because it had a quasi-experimental design using a single-arm trial and external control, the evidence level may be inferior to that of RCTs. Some guidelines recommend the use of external control groups if traditional RCTs are not feasible, particularly for rare diseases or those with high unmet medical needs.^37,38 However, few single-arm and external-control-group-design studies have been performed on dementia prevention, and the results of the current study require cautionary interpretation. Second, since this study was conducted at two Japanese sites, the feasibility of multidomain intervention classes for dementia prevention administered by non-professional instructors cannot be generalized. Meanwhile, the feasibility and effectiveness of peer education programs have been reported in other areas (e.g., basic life support and smoking cessation instruction) and may become more prevalent.^12,39 It is necessary to develop more detailed materials and guidelines for instructors in the coming years and to validate these resources in larger studies. Third, the research secretariat participated in all group classes to optimize the fidelity of the instructors’ interventions and to ensure the participants’ safety. This may have resulted in an observer effect on the instructors’ interventions, potentially influencing the feasibility and effectiveness of the outcomes. It may be possible to implement classroom programs led solely by non-professional instructors through stricter inclusion criteria, more sophisticated training programs, or the use of information and communication technology; however, this remains a future challenge. Fourth, we did not obtain certain outcomes that could have influenced the results, such as data on adherence to the self-guided multidomain lifestyle interventions.

In conclusion, this study demonstrated the feasibility and effectiveness of a 12-month multidomain intervention administered by non-experts using the MCI Handbook. The result may provide implications for the application of dementia prevention programs in community centers or older adults’ day-care, as well as implications for dementia prevention policy in ageing societies. Further research is needed to assess the long-term effects and to establish effective delivery methods for the social implementation of multidomain interventions for dementia prevention.

Supplemental Material

sj-xlsx-1-alz-10.1177_13872877251372957 - Supplemental material for Feasibility and effectiveness of a handbook-based multidomain lifestyle intervention for older adults with mild cognitive impairment: A quasi-experimental design with propensity score matching

Supplemental material, sj-xlsx-1-alz-10.1177_13872877251372957 for Feasibility and effectiveness of a handbook-based multidomain lifestyle intervention for older adults with mild cognitive impairment: A quasi-experimental design with propensity score matching by Kosuke Fujita, Fumie Kinoshita, Yujiro Kuroda, Taiki Sugimoto, Mari Yamashita, Hana Sakurai, Daichi Yamashiro, Kozaburo Shimizu, Nanae Matsumoto, Kazuaki Uchida, Yoko Yokoyama, Ayaka Onoyama, Yumiko Kobayashi, Hiroyuki Suzuki, Yoshinori Fujiwara, Hidenori Arai and Takashi Sakurai in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

The authors thank the MCI Handbook Intervention Study Office (Saeko Omura, Sae Hirai, Yu Ryu, Yasuka Sumimoto, Seiko Ito, Moe Kitago, Kyoko Saeki, Kanna Fujita, and Satoshi Nakamura) for their support. Hiroyuki Shimada, Rei Otsuka, Minoru Yamada, and Aya Seike supported the creation of the MCI Handbook. The J-MINT investigators provided the external control data. The authors also thank the BioBank at the National Center for Geriatrics and Gerontology for the quality control of the genomic data.

ORCID iD

Kosuke Fujita

Ethical considerations

All procedures involved in this study were reviewed and approved by the ethics committee of NCGG (approval no. 1603).

Consent to participate

A written informed consent form was provided to the participants who met the criteria and agreed to participate in the study. The purpose, properties, and potential risks of the study were thoroughly explained, and written consent to participate was obtained before the intervention was initiated.

Consent for publication

Not applicable

Author contributions

Kosuke Fujita: Conceptualization; Data curation; Investigation; Project administration; Validation; Visualization; Writing - original draft.

Fumie Kinoshita: Formal analysis; Methodology; Software; Writing - review & editing.

Yujiro Kuroda: Conceptualization; Data curation; Investigation; Writing - review & editing.

Taiki Sugimoto: Investigation; Methodology.

Mari Yamashita: Conceptualization; Data curation; Investigation; Project administration.

Hana Sakurai: Data curation; Investigation; Project administration.

Daichi Yamashiro: Data curation; Investigation.

Kozaburo Shimizu: Data curation; Investigation.

Nanae Matsumoto: Data curation; Investigation.

Kazuaki Uchida: Data curation; Investigation.

Yoko Yokoyama: Data curation; Investigation.

Ayaka Onoyama: Data curation; Investigation.

Yumiko Kobayashi: Formal analysis; Methodology; Software.

Hiroyuki Suzuki: Data curation; Resources.

Yoshinori Fujiwara: Conceptualization; Resources; Supervision.

Hidenori Arai: Conceptualization; Funding acquisition; Resources; Supervision.

Takashi Sakurai: Conceptualization; Funding acquisition; Methodology; Project administration; Resources; Supervision; Writing - review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by the Health Labour Sciences Research Grant (grant number 21GB1003) from the Ministry of Health Labour and Welfare, Japan Agency for Medical Research and Development under grant number JP19de0107002, and the Research Funding for Longevity Sciences (grant number 22-23) from the NCGG. The funders played no role in the study design, data collection and analysis, or manuscript preparation.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data supporting the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Supplemental material

Supplemental material for this article is available online.

References

Alzheimer’s Association. 2024 Alzheimer’s disease facts and figures. Alzheimers Dement 2024; 20: 3708–3821.

van Dyck

Swanson

Aisen

, et al. Lecanemab in early Alzheimer’s disease. N Engl J Med 2023; 388: 9–21.

Cummings

Apostolova

Rabinovici

, et al. Lecanemab: appropriate use recommendations. J Prev Alzheimers Dis 2023; 10: 362–377.

Mattap

Mohan

McGrattan

, et al. The economic burden of dementia in low- and middle-income countries (LMICs): a systematic review. BMJ Glob Health 2022; 7: e007409.

Chowdhary

Barbui

Anstey

, et al. Reducing the risk of cognitive decline and dementia: WHO recommendations. Front Neurol 2022; 12: 765584.

Kivipelto

Mangialasche

Snyder

, et al. World-Wide FINGERS network: a global approach to risk reduction and prevention of dementia. Alzheimers Dement 2020; 16: 1078–1094.

Livingston

Huntley

Liu

, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing commission. Lancet 2024; 404: 572–628.

Ngandu

Lehtisalo

Solomon

, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. Lancet 2015; 385: 2255–2263.

Wimo

Handels

Antikainen

, et al. Dementia prevention: the potential long-term cost-effectiveness of the FINGER prevention program. Alzheimers Dement 2023; 19: 999–1008.

10.

Sakurai

Sugimoto

Akatsu

, et al. Japan-multimodal intervention trial for the prevention of dementia: a randomized controlled trial. Alzheimers Dement 2024; 20: 3918–3930.

11.

Resnicow

Reddy

James

, et al. Comparison of two school-based smoking prevention programs among South African high school students: results of a randomized trial. Ann Behav Med 2008; 36: 231–243.

12.

Golechha

. Health promotion methods for smoking prevention and cessation: a comprehensive review of effectiveness and the way forward. Int J Prev Med 2016; 7: 7.

13.

Kuroda

Goto

Sugimoto

, et al. Participatory approaches for developing a practical handbook integrating health information for supporting individuals with mild cognitive impairment and their families. Health Expect 2024; 27: e13870.

14.

Boutron

Altman

Moher

, et al. CONSORT Statement for randomized trials of nonpharmacologic treatments: a 2017 update and a consort extension for nonpharmacologic trial abstracts. Ann Intern Med 2017; 167: 40–47.

15.

Reeves

Gaus

. Guidelines for reporting non-randomised studies. Forsch Komplementarmed Klass Naturheilkd 2004; 11: 46–52.

16.

Fujiwara

Suzuki

Yasunaga

, et al. Brief screening tool for mild cognitive impairment in older Japanese: validation of the Japanese version of the Montreal cognitive assessment. Geriatr Gerontol Int 2010; 10: 225–232.

17.

Perneczky

Wagenpfeil

Komossa

, et al. Mapping scores onto stages: mini-mental state examination and clinical dementia rating. Am J Geriatr Psychiatry 2006; 14: 139–144.

18.

Nara

Sugie

Takahashi

, et al. Japanese Version of the Montreal cognitive assessment cut-off score to clarify improvement of mild cognitive impairment after exercise training in community-dwelling older adults. Geriatr Gerontol Int 2018; 18: 833–838.

19.

Nelson

Rejeski

Blair

, et al. Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc 2007; 39: 1435–1445.

20.

Attkisson

Zwick

. The client satisfaction questionnaire. Psychometric properties and correlations with service utilization and psychotherapy outcome. Eval Program Plann 1982; 5: 233–237.

21.

Common Terminology Criteria for Adverse Events (CTCAE) | Protocol Development | CTEP. Accessed November 18, 2024. https://ctep.cancer.gov/protocoldevelopment/electronic_applications/ctc.htm

22.

Fujiwara

Suzuki

Yasunaga

, et al. Brief screening tool for mild cognitive impairment in older Japanese: validation of the Japanese version of the Montreal cognitive assessment. Geriatr Gerontol Int 2010; 10: 225–232.

23.

Hachisuka

Ogata

Ohkuma

, et al. Test-retest and inter-method reliability of the self-rating Barthel Index. Clin Rehabil 1997; 11: 28–35.

24.

Lawton

Brody

. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist 1969; 9: 179–186.

25.

Toba

Okochi

Takahashi

, et al. Development of a portable fall risk index for elderly people living in the community. Nihon Ronen Igakkai Zasshi 2005; 42: 346–352.

26.

Rubenstein

Harker

Salvà

, et al. Screening for undernutrition in geriatric practice: developing the short-form mini-nutritional assessment (MNA-SF). J Gerontol A Biol Sci Med Sci 2001; 56: M366–M372.

27.

Shin

Park

Lee

, et al. Usefulness of the 15-item geriatric depression scale (GDS-15) for classifying minor and major depressive disorders among community-dwelling elders. J Affect Disord 2019; 259: 370–375.

28.

Marten

Brand

Greiner

. Feasibility of the EQ-5D in the elderly population: a systematic review of the literature. Qual Life Res Int J 2022; 31: 1621–1637.

29.

Yokoro

Otaki

Imamura

, et al. Association between social network and dietary variety among community-dwelling older adults. Public Health Nutr 2023; 26: 2441–2449.

30.

Doi

Minowa

Uchiyama

, et al. Psychometric assessment of subjective sleep quality using the Japanese version of the Pittsburgh Sleep Quality Index (PSQI-J) in psychiatric disordered and control subjects. Psychiatry Res 2000; 97: 165–172.

31.

Ngandu

Lehtisalo

Korkki

, et al. The effect of adherence on cognition in a multidomain lifestyle intervention (FINGER). Alzheimers Dement 2022; 18: 1325–1334.

32.

Hawley-Hague

Horne

Campbell

, et al. Multiple levels of influence on older adults’ attendance and adherence to community exercise classes. Gerontologist 2014; 54: 599–610.

33.

Coley

Ngandu

Lehtisalo

, et al. Adherence to multidomain interventions for dementia prevention: data from the FINGER and MAPT trials. Alzheimers Dement 2019; 15: 729–741.

34.

Kebede

Ozolins

Holst

, et al. Digital engagement of older adults: scoping review. J Med Internet Res 2022; 24: e40192.

35.

Lindvall

Abzhandadze

Quinn

, et al. Is the difference real, is the difference relevant: the minimal detectable and clinically important changes in the Montreal cognitive assessment. Cereb Circ Cogn Behav 2024; 6: 100222.

36.

Nandi

Counts

Bröker

, et al. Cost of care for Alzheimer’s disease and related dementias in the United States: 2016 to 2060. NPJ Aging 2024; 10: 13.

37.

Oesterheld

Ferguson

Kraveka

, et al. Eflornithine as postimmunotherapy maintenance in high-risk neuroblastoma: externally controlled, propensity score–matched survival outcome comparisons. J Clin Oncol 2024; 42: 90–102.

38.

Velummailum

McKibbon

Brenner

, et al. Data challenges for externally controlled trials: viewpoint. J Med Internet Res 2023; 25: e43484.

39.

Eng

SWO

Ting

JXR

, et al. Global prevalence of basic life support training: a systematic review and meta-analysis. Resuscitation 2023; 186: 109771.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB