LETHE-AT: a Personalized Multidomain Lifestyle Intervention for Individuals at Increased Risk of Memory Impairment.

Overview

The goal of this clinical trial is to learn whether a hybrid multidomain lifestyle program can prevent cognitive decline and reduce dementia risk in community-dwelling adults in mid- to late life who are at increased risk of Alzheimer´s disease or related dementias but do not yet have significant cognitive impairment. The main question the study aims to answer are: * Whether the structured hybrid multidomain lifestyle intervention is feasible (e.g., adherence and retention rate), and how well the digital components are accepted and implemented in the intervention group. * Does the intervention reduces the overall burden of modifiable dementia risk factors and improves global cognitive performance compared with usual care. Researchers will compare participants assigned to the tailored hybrid multidomain lifestyle intervention group with those in a self-guided multimodal lifestyle advice group. Participants assigned to the intervention group will receive a plan adjusted to their individual dementia risk profile. A physician trained in motivational interviewing will review their progress continuously. The self-guided multimodal lifestyle advice group will receive rigid but comprehensible lifestyle health advice with reduced access to digital support tools. Participants will: * Complete an initial risk assessment that uses machine-learning triage to identify and prioritize their most important modifiable dementia risk factors. * Receive personalized recommendations for gradual lifestyle change, including physical activity, nutrition, cognitive training, other dementia risk-factor management (e.g. hearing impairment), stress \& sleep management, and social activities. * Use a smartphone and smartwatch to passively collect digital biomarkers and to complete questionnaires at regular intervals, so that physicians trained in motivational interviewing can adapt goals through shared decision making. * Use a study app as the central access point for the program, including educational content, progress tracking, and gamified challenges with social comparison and incentives.

LETHE AT is an 18 month multicentre randomized controlled trial in Austria that evaluates a tailored hybrid multidomain lifestyle intervention programme in older adults with multiple dementia risk factors and is conducted at the Medical Universities of Vienna, Innsbruck, and Graz. Previous multidomain lifestyle trials have shown that structured interventions targeting modifiable risk factors can attenuate cognitive decline and reduce dementia risk. These programme have used different delivery formats, including intensive face to face interventions (FINGER), fully digital approaches (Maintain Your Brain), and hybrid models (SMART and LETHE EU). LETHE-AT builds on this foundation by combining refined risk profiling and motivational interviewing (MI) with a centrally organized online and telephone prescreening process. This approach identifies cognitively unimpaired adults with a high burden of dementia risk factors and reduces avoidable on-site screening failures and participant disappointment. Following the prescreening, comprehensive on-site screening is conducted at three Austrian memory clinics to enroll 300 participants. Eligibility is determined using predefined inclusion and exclusion criteria, including age 55-75 years, preserved cognition, subjective cognitive decline or a first-degree family history of dementia, the presence of at least three of 14 established modifiable dementia risk factors (e.g. hearing impairment, elevated LDL cholesterol), and sufficient motivation for lifestyle change as evaluated by physicians trained in MI. Eligible participants are invited for a baseline visit, where dementia risk is quantified using LIBRA2 and ANU-ADRI, and a harmonized neuropsychological and functional assessment battery is administered by trained neuropsychologists. The study population is then stratified by sex, age and study site and randomly assigned in a 1:1 ratio to the tailored hybrid multidomain lifestyle intervention group or to the self-guided multimodal lifestyle advice group. At a subsequent in-person visit, all participants attend a structured onboarding session. In both groups, the smartphone, smartwatch and LETHE-App are introduced, standardized usage instructions are provided, and access to ongoing technical support is ensured. In the MI-guided hybrid multidomain lifestyle intervention group, onboarding additionally includes an individual consultation with a trained MI-coach. The 14 modifiable risk factors are mapped onto six lifestyle domains (diet, physical activity, cognitive engagement, sleep and stress, social interactions, and other risk-factor management) to guide content and provide a replicable coaching framework. Additionally, the modifiable risk profile is processed by a dedicated AI-based decision-support tool that generates tailored goal recommendations, and a complemental questionnaire captures personal preferences and constraints. Based on this information, participants and the MI coach select personal goals from a predefined list, with each goal allocated to one of the six lifestyle domains. The intervention programme begins with a small set of high-priority goals and is gradually expanded, with the number and intensity of goals adapted over time to each participant's risk profile, capacity and circumstances. The intervention programme is harmonized across study sites and supported by manuals to ensure consistent delivery. The LETHE-App is provided to both study arms and serves several purposes. First, it enables continuous assessment of digital markers (e.g., log-in patterns), complemented by digital biomarkers from the smartwatch (e.g., sleeping-time). Second, it delivers frequent questionnaires on system satisfaction and risk-factor status. In addition, the app offers reminders and review functions for personal goals and provides evidence-based lifestyle information for each domain, including content supplied by the Austrian Agency for Health and Food Safety (AGES). Further functions, such as gamification elements to support adherence and motivation to change, are planned for iterative development. Participants attend in-person study visits at baseline, month 9 and month 18, where they undergo an extended neuropsychological test battery, standardized dementia risk assessment and blood sampling. Brain MRI is performed as part of an imaging sub-study, and blood samples are stored for analysis of conventional risk markers and exploratory dementia-related biomarkers. Outcome assessors for cognition, imaging and laboratory measures are blinded to group allocation, and group assignment is not revealed in study documentation used for these assessments. Intervention study visits consist of individual MI-based tele-health sessions of approximately 20 to 30 minutes, scheduled at roughly three-month intervals to review progress, troubleshoot barriers and adjust goals. A harmonized protocol is used across sites, and sessions focus on strengthening intrinsic motivation for lifestyle change by exploring ambivalence, linking behaviour change to personally meaningful values and reinforcing self-efficacy in initiating and maintaining lifestyle modifications. Safety is monitored through brief adverse event questionnaires at scheduled time points and through spontaneous reports during visits or phone contacts. All study related events are documented and managed according to protocol, and the intervention is nonpharmacologic and consistent with current dementia prevention guidelines. Data are pseudonymized and stored on secure servers with restricted access and audit trails, and trial conduct follows Good Clinical Practice and standardized operating procedures with trained staff and a central data management plan. The planned sample of 300 participants reflects the focus on feasibility outcomes such as counseling acceptability, adherence and retention. Analyses follow the intention to treat principle and use linear mixed effects models with group, time and their interaction adjusted for baseline values and prespecified covariates. Effect estimates are reported with 95 percent confidence intervals, and missing data are examined in planned sensitivity analyses. No formal interim efficacy analysis is planned.