A Multidomain Intervention for Modifying Lifestyle Habits Reduces the Dementia Risk in Community-Dwelling Older Adults: A Single-Blinded Randomized Controlled Pilot Study

Abstract

We aimed to examine the feasibility and effectiveness of a multidomain intervention including intensive and maintenance programs for reducing the risk of dementia in at-risk older adults. Community-dwelling older adults (aged ≥60 years) without dementia but having several risk factors for dementia (N = 32; 89% female; mean age±standard deviation, 76.8±4.7 years) were assigned to three parallel programs: intensive plus maintenance (INT+MNT), intensive only (INT-only), and active control. Subjects in INT+MNT and INT-only groups participated in a 4-week intensive group-based lifestyle modification program that focused on physical activity, vascular risk factors, dietary habits, cognitive activities, and social engagement. INT+MNT participants underwent an additional 20-week maintenance program to consolidate modified habits. The modified Australian National University-Alzheimer’s Disease Risk Index (ANU-ADRI) score was used as the primary outcome measure for dementia risk. The changes in ANU-ADRI scores exhibited a significant group-by-time interaction: the INT+MNT group showed significant improvement at 24 weeks (β= –6.05; SE = 1.86; p = 0.002), while the INT-only group did not. Additional exploratory analyses showed that the reduction in ANU-ADRI scores was caused by changes in protective factors rather than in risk factors. The INT + MNT group also showed greater improvement in executive function at 4 and 24 weeks (both p = 0.044), whereas changes in global cognitive function did not reach significance (p = 0.055). A 24-week multidomain dementia prevention involving a maintenance strategy for sustaining modified lifestyle habits reduced the risk of dementia and improved executive function in at-risk older adults.

Keywords

Aged alzheimer’s disease dementia health promotion risk reduction behavior

INTRODUCTION

Effective modalities for dementia prevention are urgently needed. One-third of Alzheimer’s disease (AD) cases worldwide are likely attributable to modifiable risk factors [1, 2], which are thus appropriate targets for dementia prevention. Most clinical trials for dementia prevention in the past three decades have targeted a single domain, such as specific cognitive training or exercise; however, effectiveness has been inconsistent among the targeted modalities [3]. Recently, multidomain interventions have been suggested as a more useful strategy for reducing the risk of dementia by addressing the multifactorial etiology of dementia [3 –5]. However, inconclusive findings from a range of large multidomain intervention trials have been reported: an intensive protocol utilized by the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) showed improved cognitive function following a 2-year multidomain intervention [6], whereas the intermediate-intensity French Multidomain Alzheimer Preventive Trial (MAPT) and the lower-intensity Dutch Prevention of Dementia by Intensive Vascular Care (PreDIVA) identified negative cognitive outcomes [7, 8]. These studies suggest that higher intensity multidomain interventions might be more effective in preventing dementia. However, feasibility must also be considered; dementia prevention is usually performed in a community setting.

This study investigated the efficacy of a multidomain intervention for dementia risk reduction in community-dwelling at-risk, but non-demented, older adults. We developed a 24-week intermediate-intensity protocol (similar to the MAPT, but of shorter duration) that considered feasibility, sustainability, and cost-effectiveness [5]. The protocol included a 4-week intensive program followed by a 20-week maintenance program. We then investigated the feasibility and effectiveness of the protocol for reducing dementia risk scores. The participants were allocated randomly into three groups: 1) intensive intervention plus a maintenance program (INT+MNT group), 2) intensive program only (INT-only group), and 3) active control (control group). We compared the adverse events and withdrawals among those groups, and the effectiveness of our protocol was examined after 4-week intensive and 20-week maintenance program, respectively.

MATERIALS AND METHODS

Participants

Participants were recruited at a community center for dementia (CCD) between November 2016 and January 2017. The CCDs provides public services regarding prevention, early detection, and rehabilitation of dementia for community-dwelling older adults; a total of 256 CCDs has been established all over the country in Korea. One of the main roles of the CCDs is to screen dementia among all older adults aged 60 years and older living in the relevant district. Some older people voluntarily visit the CCDs, while the staffs of the CCDs also invite them by mail and telephone, or even visit them individually as an outreach service. We enrolled the participants among individuals who had received the screening procedure conducted at a CCD in Seoul by reviewing the records; the eligible subjects were contacted by telephone.

Face-to-face interviews were performed to review inclusion and exclusion criteria after attaining written informed consent. The study included subjects aged ≥60 years without cognitive impairment as determined by age-, sex-, and education-adjusted z-scores on the Mini-Mental State Examination (MMSE) [9] of >– 1.5 SD. All had risk factors for dementia including at least two of following in the past year: 1) no or low level of moderate exercise (e.g., fast walking for 30; min less than three times a week); 2) reading books, newspapers or magazines less than once a week; 3) meeting friends, relatives, or close acquaintances less than once a week; and 4) having depressed mood or low volition. Exclusion criteria were previous diagnosis of dementia or other neurological disorders; stroke in the past year or having significant sequelae of a past stroke; major psychiatric disorders in the past year; and having an absolute contraindication for participation in an exercise program. In addition, the present study excluded subjects who were participants in other exercise or cognitive training in the CCD. As our current study was a pilot study evaluating the risk reduction effects of both intensive and maintenance programs, the sample size was not based on a rigorous power analysis, but rather on the numbers in previous pilot studies investigating the effectiveness of non-pharmacological interventions for lifestyle modification (10 to 15 individuals per group).

This study was approved by the Institutional Review Board of Seoul National University Hospital and was registered in a clinical trial registry (registration NCT03786510).

Randomization and blinding

The participants were assigned randomly to two intervention groups and one control group in a 1:1:1 ratio. Stratified randomization was used for treatment allocation. Participants were first grouped into strata according to age (60–69, 70–79, and ≥80 years) and sex. Then, within each stratum participants were assigned to one of the three treatment groups using a web-based computer-generated allocation schedule by an independent statistician. This study used a single-blind design, and all assessments were performed by researchers blinded to the group allocation.

Study design and procedures

Our 24-week protocol of multidomain dementia prevention was designed to be practical in a community-based setting. Therefore, all CCD staff, including nurses, physiotherapists, and social workers, served as study assistants. The study assistants were allocated to three or four participants to check the performance of weekly tasks, provide encouragement, and monitor adverse events during all sessions. The research team trained all CCD staff to perform study procedures in a uniform manner.

Interventions

The control group received personalized advice on lifestyle modifications to prevent dementia and the usual regular 3-month health check-ups recommended by the CCD. The intervention groups (INT + MNT and INT-only) participated in a 4-week group-based intensive program; then, the INT + MNT group participated in an additional 20-week maintenance program. The protocols are detailed below.

Intensive program

The 4-week group-based intensive program was comprised of eight sessions. Each session consisted of 1) a 30 min review of at-home tasks, 2) a 50 min training period, and 3) a 10 min meeting to assign home tasks on an individual basis (Fig. 1).

Fig.1

Protocol for 4-week intensive program.

The sessions began with individual reviews of home tasks set in the previous session, and the extent of daily walking was also checked. Each participant wore a Yamax SW-200 pedometer (Yamax Corporation, Tokyo, Japan), which was validated in a previous study [10, 11], and the daily step goals were gradually increased based on the number of steps completed in the previous week. Next, the study nurse performed education and training aimed at modifying vascular risk factors (session #1), diet (#3), and cognitive and social activities (#5); then, participants made a plan for sustaining the modified lifestyle habits (#7). In even-numbered sessions (#2, 4, 6, 8), a physiotherapist conducted exercise training that consisted mainly of muscle strengthening and stretching. Finally, home tasks were assigned based on the weekly themes (e.g., cognitive and social activities); Figure 1 shows the details of the tasks. Although the participants performed the intensive program as a group, protocol implementation was individually guided by study assistants. For example, assignment of home tasks supporting cognitive activities was based on the participants’ usual habits, such as watching the news on TV with a grandson every day or attending a poetry class in the church once a week. Participants used a workbook to record their home task performance.

Maintenance program

The goal of the maintenance program was to sustain the lifestyle habits modified by the prior intensive program. The allocated study assistants continued to monitor the participants through the maintenance program. They checked the workbooks monthly to determine whether the participants adhered to the modified habits; if not, the study assistants discussed with them about why they did not, and how they could re-attempt the lifestyle tasks. Almost all sessions were conducted using face-to-face meetings; only a few sessions were done by telephone if face-to-face meetings were not possible.

Outcome assessment and measures

Blinded study nurses evaluated the patients’ demographics, anthropometrics, physical fitness, and risk factors for dementia at baseline, 4 weeks, and 24 weeks, and psychologists performed the neuropsychological tests at the same time points. A study physician (geriatric psychiatrist) evaluated cognitive status and performed physical and neurologic examinations at baseline and at 24 weeks.

Primary outcome

We used a modified version of the Australian National University-Alzheimer’s Disease Risk Index (ANU-ADRI) score as the primary outcome measure. The ANU-ADRI was developed based on a review and meta-analysis of the literatures on modifiable risk and protective factors for AD. The ANU-ADRI is a validated instrument assessing individual risk profiles for dementia that combines the odds ratios of risk factors to derive a risk score [12]. Of the 11 risk factors and 4 protective factors included in the original ANU-ADRI, we used 8 risk factors and 4 protective factors that were confirmed as useful predictors for dementia in three independent middle-aged cohorts [13]. In this study, the risk factors were age, sex, low education, diabetes, history of traumatic brain injury, smoking, depression, and low social engagement; the protective factors were light-to-moderate alcohol consumption, high physical activity, high cognitive activity, and fish consumption three or more times a week. The total scores ranged from –18 to 63 for males, and from –18 to 66 for females; higher scores indicated greater risk for AD. More detailed description of the ANU-ADRI is provided in the Supplementary Table 1.

Secondary outcome

Secondary outcomes covered five domains: physical, cognitive, functional status, quality of life, and motivation for preventing dementia. We used the Short Physical Performance Battery (SPPB) to evaluate physical function [14]. Global cognitive function was reflected by the total score on the neuropsychological battery created by the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD-TS) [15, 16]; executive function was assessed using the Color–Word Stroop Test (CWST) [17, 18]; daily functional status was evaluated using the Seoul-Instrumental Activities of Daily Living (S-IADL) assessment [19]; quality of life was assessed using the EuroQol five-dimension questionnaire (EQ-5D) [20]; and beliefs and attitudes toward behavioral changes for preventing dementia were measured using the Motivation to Change Lifestyle and Health Behaviours for Dementia Risk Reduction (MCLHB-DRR) scale [21].

Statistical analyses

Baseline characteristics were compared between groups using analysis of variance (ANOVA) for continuous variables and chi-square test for categorical variables. We used a modified intent-to-treat approach, including all randomly assigned participants with at least one post-baseline observation. Linear mixed models with factors of intervention group (reference: active control), time (reference: baseline), and their interaction were used to examine group differences in changes in primary and secondary outcomes, from baseline to the 4- and 24-week follow-ups. Additionally, we conducted two-way mixed ANOVA, where the missing data of withdrawn subjects were conservatively input as ‘no change’ by carrying forward baseline values. For post-hoc analyses, cross-sectional comparisons were conducted between groups using ANOVA (inter-group analyses); baseline ANU-ADRI was included as a covariate in analyses of primary outcomes. In addition, time-related changes within each group were examined by paired t-tests (intra-group analyses). All statistical analyses were performed in SPSS software (ver. 25.0; SPSS Inc., Chicago, IL, USA). Two-sided tests were used, with the level of significance set at p < 0.05.

RESULTS

A total of 52 individuals were assessed for eligibility after reviewing the CCD records of those who were recently screened for dementia. Among them, 14 were excluded because they did not meet our criteria and six declined to participate in the study. Therefore, a total of 32 subjects were enrolled in the trial and randomized after the baseline assessment. Six subjects withdrew during the study period and did not attend post-baseline assessments (Fig. 2). The number of such subjects did not differ significantly among the three groups (INT+MNT, 15.4%; INT-only, 22.2%; active control, 20.0%; p = 0.915). The average attendance rates of each intensive intervention session during 4 weeks were 90.0% (INT+MNT, 90.0%; INT-only, 89.7%), and most of the participants completed the homework for each session. No participant in INT+MNT group missed the maintenance session every month during 20 weeks. No adverse events were reported during the study period.

Fig.2

Flowchart for enrollment, allocation, and participation.

Table 1 shows the baseline characteristics of participants according to the group allocation. All participants had a mean (SD) age of 76.8 (4.7) years, 8.9 (4.7) years of education, and a mean MMSE score of 26.9 (3.2); 88.5% were female. No significant differences were detected among the three groups in demographic characteristics or primary and secondary outcome variables at baseline. However, post-hoc power analyses revealed that the statistical power to detect group differences on the baseline characteristics ranged from 0.07 to 0.44 indicating our analyses were underpowered due to the small sample size.

Table 1

Baseline characteristics of participants

	Active control (n = 8)	Intensive+Maintenance program (n = 11)	Intensive program only (n = 7)	p^*
Demographic variables
Age, y (SD)	76.13 (4.12)	77.36 (4.76)	76.57 (5.74)	0.854
Female, n (%)	7 (87.50)	10 (90.91)	6 (85.71)	0.947
Education, y (SD)	8.50 (4.78)	7.27 (4.92)	12.00 (3.22)	0.111
MMSE, scores (SD)	26.88 (3.80)	26.45 (3.64)	27.57 (1.27)	0.779
Primary outcome variables
^aANU-ADRI risk score, scores (SD)	17.63 (7.67)	23.64 (9.84)	15.57 (8.66)	0.153
ANU-ADRI risk factors, scores (SD)	26.00 (7.13)	29.91 (8.25)	24.43 (8.89)	0.346
ANU-ADRI protective factors, scores (SD)	–8.38 (3.20)	–6.27 (3.52)	–8.86 (2.80)	0.211
Secondary outcome variables
Physical status
SPPB, scores (SD)	9.38 (2.50)	9.82 (2.18)	9.00 (2.38)	0.766
Cognitive status
CERAD-NP total, scores (SD)	60.38 (14.32)	57.00 (10.95)	61.86 (11.48)	0.688
CWST, scores (SD)	33.43 (10.81)	19.91 (15.97)	33.86 (21.59)	0.145
Functional status
S-IADL, scores (SD)	2.50 (1.41)	3.09 (2.51)	3.86 (2.55)	0.514
Quality of life
EQ-5D, scores (SD)	7.00 (1.41)	6.45 (1.51)	5.57 (0.79)	0.134
Dementia literacy and motivation for risk reduction
MCLHB-DRR, scores (SD)	80.13 (16.43)	88.90 (9.47)	86.57 (6.75)	0.289

ANU-ADRI, Australian National University-Alzheimer’s disease risk index; SPPB, short physical performance battery; CERAD-NP, consortium to establish a registry for Alzheimer’s disease–neuropsychological battery; CWST, color word Stroop test; S-IADL, Seoul instrumental activities of daily living; EQ-5D, EuroQol- 5 dimension; MCLHB-DRR, motivation to change lifestyle and health behaviors for dementia risk reduction scale. Note: Higher scores indicate higher risks for dementia (ANU-ADRI, from -18 to 63 [men] or 66 [women]); better physical function (SPPB, from 0 to 12); better global cognitive function (CERAD-NP total score, from 0 to 100); better executive function (CWST, no defined range of scores; the number of items properly named in 45 sec ranges from 0 to 65 in our sample); poorer instrumental daily function (S-IADL, from 0 to 45); to feel highly susceptible to dementia, but, be more ready to change (MCLHB-DRR, from 27 to 135).^*ANOVA for continuous variables; chi-square tests for categorical variables ^aANU-ADRI risk factors include age, sex, low education, diabetes, history of traumatic brain injury, smoking, depression, and low social engagement; ANU-ADRI protective factors include light-to-moderate alcohol consumption, high physical activity, high cognitive activity, and fish consumption (≥3 times per week).

Mixed effect model analyses showed a significant group-by-time interaction for the primary outcome measure (ANU-ADRI score). The INT+MNT group showed a significant improvement in the ANU-ADRI score at 24 weeks (Table 2), whereas no significant difference was observed at 4 weeks among the three groups. The group-by-time interaction results were similar in two-way mixed ANOVA, in which missing follow-up data were conservatively substituted with baseline values, and the analysis was adjusted for baseline ANU-ADRI (F [3.21, 36.88] = 3.319, p = 0.028). Post-hoc within group analyses revealed that the pattern of change in the ANU-ADRI score over the three measurement occasions differed across the three groups. The INT+MNT group showed a significant improvement in the ANU-ADRI score at week 4 and the decreased level continued until 24 weeks, whereas the INT-only and the active control showed a significant increase in the ANU-ADRI score at 24 weeks even though they decreased at 4 weeks (Fig. 3).

Table 2

Results from the analyses using linear mixed effect models for the differences of primary and secondary outcomes

	Primary outcome			Secondary outcome
	ANU-ADRI	ANU-ADRI risk factor	ANU-ADRI protective factor	SPPB	CERAD-NP total score	CWST	S-IADL	EQ-5D	MCLHB-DRR
	β (SE)^a			β (SE)^a
Constant	17.63 (3.19)	26.00 (2.86)	–8.38 (1.11)	9.38 (0.76)	60.38 (4.53)	34.23 (6.01)	2.50 (0.67)	7.00 (0.40)	80.13 (4.22)
intervention group^b
INT+MNT	6.01 (4.20)	3.91 (3.76)	2.10 (1.46)	0.44 (0.99)	–3.38 (5.95)	–14.32 (7.70)	0.59 (0.88)	–0.55 (0.53)	8.93 (5.65)
INT-only	–2.05 (4.68)	–1.57 (4.19)	–0.48 (1.62)	–0.38 (1.11)	1.48 (6.63)	–0.37 (8.51)	1.36 (0.98)	–1.43 (0.59)	6.45 (6.17)
Time^c
4 weeks	–1.25 (1.88)	–0.13 (1.53)	–1.13 (1.37)	0.75 (0.89)	3.00 (2.87)	–3.73 (7.10)	–0.38 (0.95)	–0.38 (0.51)	–3.75 (5.67)
24 weeks	1.95 (1.41)	1.15 (1.14)	0.73 (1.18)	0.61 (0.74)	1.63 (2.17)	–9.70 (5.87)	0.49 (0.97)	–0.32 (0.44)	–9.94 (5.16)
Interaction of
group by time
4 week - INT+MNT	–2.30 (2.46)	0.31 (2.01)	–2.60 (1.81)	–0.20 (1.17)	3.64 (3.78)	18.91 (9.17) ^*	–1.44 (1.25)	–0.26 (0.66)	–3.03 (7.53)
24 week - INT+MNT	–6.05 (1.86) ^*	–0.94 (1.52)	–4.94 (1.56) ^*	0.82 (0.98)	5.67 (2.87)	16.12 (7.71) ^*	–2.57 (1.29)	–0.67 (0.58)	–1.13 (6.94)
4 week - INT-only	–1.18 (2.74)	–0.02 (2.24)	–1.16 (2.01)	–0.18 (1.30)	–0.29 (4.21)	10.87 (10.17)	–1.05 (1.39)	1.09 (0.74)	2.89 (8.30)
24 week - INT-only	0.48 (2.02)	0.70 (1.64)	–0.16 (1.69)	0.96 (1.06)	3.23 (3.11)	1.84 (8.32)	–2.20 (1.40)	0.32 (0.63)	10.37 (7.39)

ANU-ADRI, Australian National University-Alzheimer’s disease risk index; SPPB, short physical performance battery; CERAD-NP, consortium to establish a registry for Alzheimer’s disease–neuropsychological battery; CWST, color word stroop test; S-IADL, Seoul instrumental activities of daily living; EQ-5D, EuroQol five dimensions, MCLHB-DRR, motivation to change lifestyle and health behaviors for dementia risk reduction scale; SE, standard error; INT+MNT, intensive plus maintenance program; INT-only, intensive program only. Note: Higher scores indicate higher risks for dementia (ANU-ADRI, from –18 to 63 [men] or 66 [women]); better physical function (SPPB, from 0 to 12); better global cognitive function (CERAD-NP total score, from 0 to 100); better executive function (CWST, no defined range of scores; the number of items properly named in 45 sec ranges from 0 to 65 in our sample); poorer instrumental daily function (S-IADL, from 0 to 45); to feel highly susceptible to dementia, but, be more ready to change (MCLHB-DRR, from 27 to 135).^aLinear mixed model including factors of group, time, and group×time interaction ^bActive control as reference group ^cBaseline as reference time ^* Significance at p < 0.05.

Fig.3

The ANU-ADRI score over time according to intervention groups. ANU-ADRI, Australian National University-Alzheimer’s disease risk index; INT, intensive program; MNT, maintenance program. Note: Two-way mixed model analyses of covariance, in which missing data of follow-up were conservatively substituted by baseline values, included factors of group, time, and group×time interaction, and adjusting for baseline ANU-ADRI; significant group-by-time interaction (F = 3.319, p = 0.028). (a) ANU-ADRI total score, estimated mean (standard error [SE]) Active control: baseline, 17.63 (3.15); at week 4, 16.38 (3.24); at week 24, 19.38 (3.13) (at week 4 < at week 24) INT + MNT group: baseline, 23.64 (2.69); at week 4, 20.09 (2.76); at week 24, 19.82 (2.67) (baseline > at week 4; baseline > at week 24) INT-only group: baseline, 15.57 (3.37); at week 4, 13.14 (3.46); at week 24, 18.00 (3.35) (at week 4<at week 24). (b) ANU-ADRI risk score, estimated mean (SE) Active control: baseline, 26.00 (2.87); at week 4, 25.88 (2.86); at week 24, 27.00 (2.83) INT + MNT group: baseline, 29.91 (2.44); at week 4, 30.09 (2.44); at week 24, 30.09 (2.41) INT-only group: baseline, 24.43 (3.06); at week 4, 24.29 (3.06); at week 24, 26.29 (3.02). (c) ANU-ADRI protective score, estimated mean (SE) Active control: baseline, –8.38 (1.15); at week 4, –9.50 (0.99); at week 24, –7.63 (2.83) INT + MNT group: baseline, –6.27 (0.98); at week 4, –10.00 (0.84); at week 24, –10.46 (0.99) (baseline > at week 4; baseline > at week 24) INT-only group: baseline, –8.86 (1.23); at week 4, –11.13 (1.05); at week 24, –8.29 (1.24) (at week 4 < at week 24).

Categorization of the ANU-ADRI factors into risk and protective classes revealed that the intervention group by time interaction was significant only for the protective factors, with the INT + MNT group evidencing a significant difference in the ANU-ADRI protective factor score at 24 weeks (Table 2). Intra-group analyses revealed a significant improvement in the overall ANU-ADRI score in the INT+MNT group from baseline to 4 and 24 weeks, which was due to a sustained increase in protective factors at 4 weeks and 24 weeks (Supplementary Table 2; Fig. 3). Among the individual protective factors, changes in cognitive activity and fish intake were associated with sustained improvements in the INT + MNT group at 24 weeks. The INT-only group showed a significant change in ANU-ADRI protective factors only at 4 weeks, which was not sustained at 24 weeks.

Among the secondary outcomes, the INT+MNT group showed a significant improvement in the CWST score at 4 weeks, which was sustained at 24 weeks (Table 2). The INT + MNT group showed a trend toward improvement at 24 weeks in the CERAD-TS (p = 0.055). Similar results were obtained for the S-IADL score (p = 0.053). No other significant differences were observed between intervention groups over time.

DISCUSSION

In this pilot study, the dementia risk score was significantly decreased in the INT + MNT group at 24 weeks compared to the INT-only and control groups, whereas no significant differences were observed in risk score changes at 4 weeks among the three groups. These results suggest that our multidomain intervention was effective only with inclusion of the maintenance program, and not with the intensive program alone. In addition, only the INT + MNT group showed improvement in executive function at 24 weeks.

To the best of our knowledge, no multidomain dementia prevention trials have examined the additive effect of a maintenance program following an intensive program for reducing the risk of dementia. The protocol used in our study was similar to that utilized in the MAPT, which incorporated a 2-month intensive multidomain intervention followed by a monthly check-up for 3 years. However, the goal of the MAPT study differed from that of our study, as they compared a combined lifestyle intervention plus N-3 polyunsaturated fatty acid supplement to each intervention alone [8]. Although the primary cognitive outcome was not improved in the MAPT, the authors reported a beneficial effect for individuals with an increased risk of dementia [8]. Our protocol, which was of intermediate intensity and shorter duration compared to the MAPT, showed a significant reduction in dementia risk score and improvement in executive function among at-risk older adults. Our findings are in line with previous trials for preventing chronic metabolic disease; it has been suggested that ongoing support is key to maintaining positive results [22].

The ANU-ADRI score was used to examine the efficacy of an online intervention for dementia risk reduction in a previous randomized controlled trial study for middle-aged adults (N = 176) [23]. Significant improvements in ANU-ADRI scores were evident in response to intervention; the Δz-scores were –0.28 (online intervention only) and –0.48 (online intervention with the addition of face-to-face sessions) [23]. Our study showed a similar degree of change on the standardizing ANU-ADRI score of Δz = –0.36 for the INT + MNT group at 24 weeks; however, the baseline score was markedly higher in our group, perhaps due to participant characteristics (older age and lower education) and the inclusion criteria used when selecting the at-risk population.

We explored the effects of intervention on ANU-ADRI scores in separate analyses of the risk and protective factors; the improvement was attributable principally to increases in protective factor scores, particularly cognitive activities and fish intake. This finding was consistent with the results of a previous study using the ANU-ADRI [23]. Thus, enhancement of protective behaviors is the easier way to reduce the dementia risk in the short term, perhaps because ANU-ADRI risk factors such as smoking and depression are relatively difficult to modify in the short term; prolonged intervention or longer follow-up may be required. Nonetheless, our goal was to devise more feasible and cost-effective protocols for dementia prevention in a community setting, and future interventional strategies should emphasize improvements in protective factors.

The INT + MNT group showed a significant difference in executive function at 24 weeks as measured by the CWST, whereas there were no significant inter-group differences in global cognitive function, as indexed by CERAD-TS. The FINGER data also revealed a significant effect of intervention on executive function and process speed, but not in the memory domain [6]. Compared to other cognitive functions, executive function declines earlier with increasing age [24]; therefore, executive function deficits might be detected more easily among relatively healthy older adults. In multidomain interventions including various lifestyle modifications, executive function might appear more affected, as it is closely related to the performance of daily activities [25].

Our pilot study established the feasibility of a 24-week multidomain intervention for dementia risk reduction in community-dwelling older adults. Although CCD staff supervised all participants in both programs, each staff member was responsible for only three or four individuals; detailed protocol manuals were available, and supervision did not require much time or effort. In addition, intervention was not associated with any adverse event, although we included participants with relatively low baseline SPPB scores (i.e., low physical function) compared to the average score of older Korean adults [26].

A major strength of our study was that the effect of the maintenance program was assessed separately from that of the intensive program. Although further studies are needed to confirm these beneficial effects with larger samples and post-intervention follow-up, our intervention-with-maintenance program appears to be feasible and effective in terms of dementia risk reduction in a community setting. Nevertheless, this study has several limitations. First, reporting bias might have occurred because the primary outcome measure was self-reported. Second, only the assessors were blinded to group allocation. It was impossible to use a double-blind design because the participants experienced the intervention protocols. Third, although the baseline scores of ANU-ADRI and CWST was not different significantly by group, the INT+MNT group showed worse scores on them at baseline; therefore, they have possibly more room for improvement compare with other groups. Lastly, the small sample size could affect the non-significant findings in INT-only group. Further trials using a larger sample size with a longer-term follow-up period should be conducted based on the positive findings of this pilot study.

Conclusion

The findings of our pilot study support the feasibility and efficacy of a multidomain intervention with maintenance program in terms of reducing the dementia risk in community-dwelling older adults at risk. Future studies with a larger sample size, longer follow-up, and additional analyses of cost-effectiveness are needed.

Footnotes

ACKNOWLEDGMENTS

This study was supported by grants from Seoul Metropolitan Center for Dementia.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Norton

, Matthews

, Barnes

, Yaffe

, Brayne

(2014) Potential for primary prevention of Alzheimer’s disease: An analysis of population-based data. Lancet Neurol 13, 788–794.

Ashby-Mitchell

, Burns

, Shaw

, Anstey

(2017) Proportion of dementia in Australia explained by common modifiable risk factors. Alzheimers Res Ther 9, 11.

Andrieu

, Coley

, Lovestone

, Aisen

, Vellas

(2015) Prevention of sporadic Alzheimer’s disease: Lessons learned from clinical trials and future directions. Lancet Neurol 14, 926–944.

Sindi

, Mangialasche

, Kivipelto

(2015) Advances in the prevention of Alzheimer’s disease. F1000Prime Rep 7, 50.

Kivipelto

, Mangialasche

, Ngandu

(2018) Lifestyle interventions to prevent cognitive impairment, dementia and Alzheimer disease. Nat Rev Neurol 14, 653–666.

Ngandu

, Lehtisalo

, Solomon

, Levalahti

, Ahtiluoto

, Antikainen

, Backman

, Hanninen

, Jula

, Laatikainen

, Lindstrom

, Mangialasche

, Paajanen

, Pajala

, Peltonen

, Rauramaa

, Stigsdotter-Neely

, Strandberg

, Tuomilehto

, Soininen

, Kivipelto

(2015) 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 385, 2255–2263.

Moll van Charante

, Richard

, Eurelings

, van Dalen

, Ligthart

, van Bussel

, Hoevenaar-Blom

, Vermeulen

, van Gool

(2016) Effectiveness of a 6-year multidomain vascular care intervention to prevent dementia (preDIVA): A cluster-randomised controlled trial. Lancet 388, 797–805.

Andrieu

, Guyonnet

, Coley

, Cantet

, Bonnefoy

, Bordes

, Bories

, Cufi

, Dantoine

, Dartigues

, Desclaux

, Gabelle

, Gasnier

, Pesce

, Sudres

, Touchon

, Robert

, Rouaud

, Legrand

, Payoux

, Caubere

, Weiner

, Carrie

, Ousset

, Vellas

(2017) Effect of long-term omega 3 polyunsaturated fatty acid supplementation with or without multidomain intervention on cognitive function in elderly adults with memory complaints (MAPT): A randomised, placebo-controlled trial. Lancet Neurol 16, 377–389.

Kim

, Jhoo

, Park

, Kim

, Ryu

, Moon

, Choo

, Lee

, Yoon

, Do

, Lee

, Kim

(2010) Korean version of mini mental status examination for dementia screening and its’ short form. Psychiatry Investig 7, 102–108.

10.

Motl

, McAuley

, Snook

, Scott

(2005) Accuracy of two electronic pedometers for measuring steps taken under controlled conditions among ambulatory individuals with multiple sclerosis. Mult Scler 11, 343–345.

11.

Kenyon

, McEvoy

, Sprod

, Maher

(2013) Validity of pedometers in people with physical disabilities: A systematic review. Arch Phys Med Rehabil 94, 1161–1170.

12.

Anstey

, Cherbuin

, Herath

(2013) Development of a new method for assessing global risk of Alzheimer’s disease for use in population health approaches to prevention. Prev Sci 14, 411–421.

13.

Anstey

, Cherbuin

, Herath

, Qiu

, Kuller

, Lopez

, Wilson

, Fratiglioni

(2014) A self-report risk index to predict occurrence of dementia in three independent cohorts of older adults: The ANU-ADRI. PLoS One 9, e86141.

14.

Guralnik

, Ferrucci

, Simonsick

, Salive

, Wallace

(1995) Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med 332, 556–561.

15.

Lee

, Lee

, Kim

, Jhoo

, Kim

, Lee

, Kim

, Han

, Woo

(2002) Development of the Korean version of the Consortium to Establish a Registry for Alzheimer’s Disease Assessment Packet (CERAD-K): Clinical and neuropsychological assessment batteries. J Gerontol B Psychol Sci Soc Sci 57, P47–53.

16.

Choe

, Sohn

, Choi

, Byun

, Seo

, Han

, Kim

, Yoon

, Lee

, Park

, Woo

, Lee

(2014) Association of homocysteine with hippocampal volume independent of cerebral amyloid and vascular burden. Neurobiol Aging 35, 1519–1525.

17.

Graf

, Uttl

, Tuokko

(1995) Color- and picture-word Stroop tests: Performance changes in old age. J Clin Exp Neuropsychol 17, 390–415.

18.

Seo

, Lee

, Choo

, Kim

, Youn

, Jhoo

, Woo

(2008) Normative study of the Stroop Color and Word Test in an educationally diverse elderly population. Int J Geriatr Psychiatry 23, 1020–1027.

19.

, Kim

, Kwon

, Kim

, Lee

, Ko

, Jo

, Kim

(2004) A study on the reliability and validity of Seoul-Instrumental Activities of Daily Living (S-IADL). J Korean Neuropsychiatr Assoc 43, 189–199.

20.

Kim

, Cho

, Uhm

, Kim

, Bae

(2005) Cross-cultural adaptation and validation of the Korean version of the EQ-5D in patients with rheumatic diseases. Qual Life Res 14, 1401–1406.

21.

Kim

, Sargent-Cox

, Cherbuin

, Anstey

(2014) Development of the motivation to change lifestyle and health behaviours for dementia risk reduction scale. Dement Geriatr Cogn Dis Extra 4, 172–183.

22.

Tapsell

, Neale

(2016) The effect of interdisciplinary interventions on risk factors for lifestyle disease: A literature review. Health Educ Behav 43, 271–285.

23.

Anstey

, Bahar-Fuchs

, Herath

, Kim

, Burns

, Rebok

, Cherbuin

(2015) Body brain life: A randomized controlled trial of an online dementia risk reduction intervention in middle-aged adults at risk of Alzheimer’s disease. Alzheimers Dement (N Y) 1, 72–80.

24.

Luszcz

(2011) Executive function and cognitive aging. In Handbook of the Psychology of Aging (Seventh Edition), Elsevier, pp. 59–72.

25.

Sapkota

, Backman

, Dixon

(2017) Executive function performance and change in aging is predicted by apolipoprotein E, intensified by catechol-O-methyltransferase and brain-derived neurotrophic factor, and moderated by age and lifestyle. Neurobiol Aging 52, 81–89.

26.

Jung

, Kim

, Ahn

, Lim

, Han

, Kim

(2014) Prevalence and outcomes of frailty in Korean elderly population: Comparisons of a multidimensional frailty index with two phenotype models. PLoS One 9, e87958.