The Internet-Based Conversational Engagement Clinical Trial (I-CONECT) in Socially Isolated Adults 75+ Years Old: Randomized Controlled Trial Protocol and COVID-19 Related Study Modifications

Mild cognitive impairment detection from facial video interviews by applying spatial-to-temporal attention module

Early detection of Mild Cognitive Impairment (MCI) leads to early interventions to slow the progression from MCI into dementia. Deep Learning (DL) algorithms could help achieve early non-invasive and low-cost detection of MCI. This paper presents the detection of MCI in older adults using DL models based only on facial features extracted from video-recorded conversations at home. We used the data collected from the I-CONECT behavioral intervention study (NCT02871921), where several sessions of semi-structured interviews between socially isolated older individuals and interviewers were video recorded. We develop a framework that extracts holistic spatial facial features using a convolutional autoencoder and temporal information using transformers. We proposed the Spatial-to-Temporal Attention Module (STAM) to detect the I-CONECT study participants' cognitive conditions (MCI vs. those with normal cognition (NC)) using facial and interaction features. The interaction features of the facial features improved the prediction performance compared with applying facial features solely. The detection accuracy using this combined method reached 88%, whereas the accuracy without applying the segments and sequences information of the facial features within a video on a certain theme was 84%. Overall, the results show that spatiotemporal facial features modeled using DL algorithms have a discriminating power for the detection of MCI.

Best practices for implementing a technology-based intervention protocol: Participant and researcher considerations

Technology-based interventions present a promising approach to support health and wellness for older adults with a range of cognitive abilities. Technology can enhance access to interventions and support scaling of programs to reach more people. However, the use of technology for intervention delivery requires particular attention to users' needs and preferences and ensuring the materials are adaptable and supportive of a diverse range of technology proficiency levels. We share best practices based on lessons learned from the deployment of a randomized controlled trial (RCT) wherein we delivered an 8-week social engagement intervention through a video technology platform called OneClick for older adults with varying cognitive abilities. We developed a set of best practices and guidelines informed by the lessons learned through this RCT implementation. Technology-based interventions require attention to the application (e.g., video calls), system requirements (e.g., system memory, broadband internet), training (e.g., adaptability based on user competency), and support (e.g., handouts, live contact). These best practices relate to user needs; training design; personnel responsibility; structuring delivery and content; and evaluating success. These research-based best practices can guide the design, development, and implementation of technology-based interventions to support older adults with varying cognitive abilities.

Methods and baseline characteristics for a social engagement technology-based randomized controlled trial for older adults

Background

Growing evidence suggests that increasing opportunities for social engagement has the potential to support successful aging. However, many older adults may have limited access to in-person social engagement opportunities due to barriers such as transportation. We outline the development, design, methodology, and baseline characteristics of a randomized controlled trial that assessed the benefits of a social engagement intervention delivered through the OneClick video conferencing platform to older adults with varying levels of cognitive functioning.

Methods

Community-dwelling older adults with and without cognitive challenges were randomly assigned to a social engagement intervention group or a waitlist control group. Participants were asked to attend twice-weekly social engagement events for 8 weeks via OneClick. Outcomes included social engagement and technology acceptance for both groups at baseline, week-4, and week-8 assessments. As an extension, the waitlist control group had an opportunity to participate in the intervention, with outcomes assessed at weeks 12 and 16.

Results

We randomly assigned 99 participants (mean age = 74.1 ± 6.7, range: 60-99), with 50 in the immediate intervention group and 49 in the waitlist control group. About half of the participants reported living alone (53.5%), with a third (31%) falling into the cognitively impaired range on global cognitive screening. The groups did not differ at baseline on any of the outcome measures.

Conclusions

Outcomes from this study will provide important information regarding the feasibility and efficacy of providing technology-based social engagement interventions to older adults with a range of cognitive abilities.

Linguistic-based Mild Cognitive Impairment detection using Informative Loss

This paper presents a deep learning method using Natural Language Processing (NLP) techniques, to distinguish between Mild Cognitive Impairment (MCI) and Normal Cognitive (NC) conditions in older adults. We propose a framework that analyzes transcripts generated from video interviews collected within the I-CONECT study project, a randomized controlled trial aimed at improving cognitive functions through video chats. Our proposed NLP framework consists of two Transformer-based modules, namely Sentence Embedding (SE) and Sentence Cross Attention (SCA). First, the SE module captures contextual relationships between words within each sentence. Subsequently, the SCA module extracts temporal features from a sequence of sentences. This feature is then used by a Multi-Layer Perceptron (MLP) for the classification of subjects into MCI or NC. To build a robust model, we propose a novel loss function, called InfoLoss, that considers the reduction in entropy by observing each sequence of sentences to ultimately enhance the classification accuracy. The results of our comprehensive model evaluation using the I-CONECT dataset show that our framework can distinguish between MCI and NC with an average area under the curve of 84.75%.

Home-Based Cognitive Intervention for Healthy Older Adults Through Asking Robots Questions: Randomized Controlled Trial

Background

Asking questions is common in conversations, and while asking questions, we need to listen carefully to what others say and consider the perspective our questions adopt. However, difficulties persist in verifying the effect of asking questions on older adults' cognitive function due to the lack of a standardized system for conducting experiments at participants' homes.

Objective

This study examined the intervention effect of cognitive training moderated by robots on healthy older adults. A focus on the feasibility of the intervention at participants' homes was also maintained. Feasibility was evaluated by considering both the dropout rate during the intervention and the number of questions posed to each participant during the experiment.

Methods

We conducted a randomized controlled trial with 81 adults older than 65 years. Participants were recruited through postal invitations and then randomized into 2 groups. The intervention group (n=40) received sessions where participants listened to photo-integrated stories and posed questions to the robots. The control group (n=41) received sessions where participants listened to photo-integrated stories and only thanked the robots for confirming participation. The participants participated in 12 dialogue sessions for 2-3 weeks. Scores of global cognitive functioning tests, recall tests, and verbal fluency tasks measured before and after the intervention were compared between the 2 groups.

Results

There was no significant intervention effect on the Telephone Interview for Cognitive Status-Japanese scores, recall tests, and verbal fluency tasks. Additionally, our study successfully concluded with no participant dropouts at follow-up, confirming the feasibility of our approach.

Conclusions

There was no statistically significant evidence indicating intervention benefits for cognitive functioning. Although the feasibility of home-based interventions was demonstrated, we identified areas for improvement in the future, such as setting up more efficient session themes. Further research is required to identify the effectiveness of an improved cognitive intervention involving the act of asking questions.

Internet-Based Conversational Engagement Randomized Controlled Clinical Trial (I-CONECT) Among Socially Isolated Adults 75+ Years Old With Normal Cognition or Mild Cognitive Impairment: Topline Results

BACKGROUND AND OBJECTIVES

Social isolation is a risk factor for cognitive decline and dementia. We conducted a randomized controlled clinical trial (RCT) of enhanced social interactions, hypothesizing that conversational interactions can stimulate brain functions among socially isolated older adults without dementia. We report topline results of this multisite RCT (Internet-based conversational engagement clinical trial [I-CONECT]; NCT02871921).

RESEARCH DESIGN AND METHODS

The experimental group received cognitively stimulating semistructured conversations with trained interviewers via internet/webcam 4 times per week for 6 months (induction) and twice per week for an additional 6 months (maintenance). The experimental and control groups both received weekly 10 minutes telephone check-ins. Protocol modifications were required due to the coronavirus disease 2019 pandemic.

RESULTS

A total of 186 participants were randomized. After the induction period, the experimental group had higher global cognitive test scores (Montreal Cognitive Assessment [primary outcome]; 1.75 points [p = .03]) compared with the control group. After induction, experimental group participants with normal cognition had higher language-based executive function (semantic fluency test [secondary outcome]; 2.56 points [p = .03]). At the end of the maintenance period, the experimental group of mild cognitive impairment subjects had higher encoding function (Craft Story immediate recall test [secondary outcome]; 2.19 points [p = .04]). Measure of emotional well-being improved in both control and experimental groups. Resting-state functional magnetic resonance imaging showed that the experimental group had increased connectivity within the dorsal attention network relative to the control group (p = .02), but the sample size was limited.

DISCUSSION AND IMPLICATIONS

Providing frequent stimulating conversational interactions via the internet could be an effective home-based dementia risk-reduction strategy against social isolation and cognitive decline.

CLINICAL TRIALS REGISTRATION NUMBER

NCT02871921.

What is the next step of ICT development? The changes of ICT use in promoting elderly healthcare access: A systematic literature review

The objective of this study was to undertake a comprehensive review of the evidence published, with a focus on understanding the experiences of the elderly in leveraging Information and Communication Technology (ICT) for their healthcare needs during the COVID-19 period. In compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, this review scrutinized all peer-reviewed articles in English sourced from PubMed, PsycINFO, Scopus, and Web of Science, targeting studies that focused exclusively on the elderly within the COVID-19 timeframe, incorporated ICT-based technology as intervention, and were associated with the assessment of the process of employing ICT for healthcare needs. The search strategy identified 1752 records, of which 34 studies met the inclusion criteria. The functionality of ICT was categorized, types of barriers were identified, and the subsequent changes that the elderly population underwent were synthesized and deliberated. This review offers valuable insights into the elderly's subjective experiences in utilizing ICT, which may offer guidance for future ICT development geared towards enhancing the well-being of the elderly. Future research should incorporate the perspectives of relevant healthcare providers in evaluating the effectiveness of ICT usage. Further studies are also needed on underserved elderly groups to provide a more holistic view.

Who Benefited Most from the Internet-Based Conversational Engagement RCT (I-CONECT)? Application of the Personalized Medicine Approach to a Behavioral Intervention Study

BACKGROUND

Many Alzheimer's Disease (AD) clinical trials have failed to demonstrate treatment efficacy on cognition. It is conceivable that a complex disease like AD may not have the same treatment effect due to many heterogeneities of disease processes and individual traits.

OBJECTIVES

We employed an individual-level treatment response (ITR) approach to determine the characteristics of treatment responders and estimated time saved in cognitive decline using the Internet-based Conversational Engagement Clinical Trial (I-CONECT) behavioral intervention study as a model.

DESIGN AND SETTING

I-CONECT is a multi-site, single-blind, randomized controlled trial aimed to improve cognitive functions through frequent conversational interactions via internet/webcam. The experimental group engaged in video chats with study staff 4 times/week for 6 months; the control group received weekly 10-minute check-in phone calls.

PARTICIPANTS

Out of 186 randomized participants, current study used 139 participants with complete information on both baseline and 6-month follow-up (73 with mild cognitive impairment (MCI), 66 with normal cognition; 64 in the experimental group, and 75 in the control group).

MEASUREMENTS

ITR scores were generated for the Montreal Cognitive Assessment (MoCA) (global cognition, primary outcome) and Category Fluency Animals (CFA) (semantic fluency, secondary outcome) that showed significant efficacy in the trial. ITR scores were generated through 300 iterations of 3-fold cross-validated random forest models. The average treatment difference (ATD) curve and the area between the curves (ABC) were estimated to measure the heterogeneity of treatment responses. Responder traits were identified using SHapley Additive exPlanations (SHAP) and decision tree models. The time saved in cognitive decline was explored to gauge clinical meaningfulness.

RESULTS

ABC statistics showed substantial heterogeneity in treatment response with MoCA but modest heterogeneity in treatment response with CFA. Age, cognitive status, time spent with family and friends, education, and personality were important characteristics that influenced treatment responses. Intervention group participants in the upper 30% of ITR scores demonstrated potential delays of 3 months in semantic fluency (CFA) and 6 months in global cognition (MoCA), assuming a 5-fold faster natural cognitive decline compared to the control group during the post-treatment period.

CONCLUSIONS

ITR-based analyses are valuable in profiling treatment responders for features that can inform future trial design and clinical practice. Reliably measuring time saved in cognitive decline is an area of ongoing research to gain insight into the clinical meaningfulness of treatment.

Including Socially Isolated Black, Older Old Adults (Aged 80 and Above) with and without Mild Cognitive Impairment in a Clinical Trial: Recruitment Strategies and Perspectives

Objective

This study describes strategies for the recruitment of socially isolated older old Black individuals to participate in the "Internet-based conversational engagement clinical trial (I-CONECT)" (Clinical Trial.gov: NCT02871921) and lessons learned in this critical population segment.

Methods

Best practice strategies to recruit the target population included mass mailings, advertisements, and direct community outreach, including the collaboration with a community group created to reach Black individuals interested in research participation. We also made protocol changes to measure recruitment criteria for older old Black adults more accurately and to increase their participation.

Results

Descriptive data related to the challenges and successes in recruiting Black participants compared to the White participants is presented. The primary site contacted 17,523 primarily White potential participants and enrolled/randomized 145 White and 2 Asian/mixed race participants (0.8%). The Midwest site contacted 12,141 Black potential participants and enrolled/randomized 39 (0.3%) participants.

Discussion

While best practices were employed, several factors complicated recruitment, including the need to adjust recruitment criteria, navigate regional regulations, and respect diverse community preferences.

Conclusion

Older old African Americans are reachable and willing to participate in research when considering their beliefs and practices, influenced by their community and experience.

Short: Causal structural learning of conversational engagement for socially isolated older adults

Social isolation has become a growing public health concern in older adults and older adults with mild cognitive impairment. Coping strategies must be developed to increase social contact for socially isolated older adults. In this paper, we explored the conversational strategy between trained conversation moderators and socially isolated adults during a conversational engagement clinical trial (Clinicaltrials.gov: NCT02871921). We carried out structural learning and causality analysis to investigate the conversation strategies used by the trained moderators to engage socially isolated adults in the conversation and the causal effects of the strategy on engagement. Causal relations and effects were inferred between participants' emotions, the dialogue strategies used by moderators, and participants' following emotions. The results found in this paper may be used to support the development of cost-efficient, trustworthy AI- and/or robot-based platform to promote conversational engagement for older adults to address the challenges in social interaction.

Randomised controlled trials for the prevention of cognitive decline or dementia: A systematic review

Dementia prevention research has progressed rapidly in recent years, with publication of several large lifestyle intervention trials, and renewed interest in pharmacological interventions, notably for individuals with Alzheimer's disease biomarkers, warranting an updated review of results and methodology. We identified 112 completed trials testing the efficacy of single-domain pharmacological (n = 33, 29%), nutritional (n = 27, 24%), physical activity (n = 18, 16%) and cognitive stimulation (n = 13, 12%), or multidomain (n = 22, 20%) interventions on incident dementia, or a relevant intermediate marker (e.g. cognitive function, biomarkers or dementia risk scores) in people without dementia. The earliest trials tested pharmacological interventions or nutritional supplements, but lifestyle interventions predominated in the last decade. In total, 21 (19%) trials demonstrated a clear beneficial effect on the pre-specified primary outcome (or all co-primary outcomes), but only two (10%) were large-scale (testing blood pressure lowering (Syst-Eur) or multidomain (FINGER) interventions on incident dementia and cognitive change in cognitive function, respectively). Of the 116 ongoing trials, 40% (n = 46) are testing multidomain interventions. Recent methodological shifts concern target populations, primary outcome measures, and intervention design, but study design remains constant (parallel group randomised controlled trial). Future trials may consider using adaptive trials or interventions, and more targeted approaches, since certain interventions may be more effective in certain subgroups of the population, and at specific times in the life-course. Efforts should also be made to increase the representativeness and diversity of prevention trial populations.

Emotional characteristics of socially isolated older adults with MCI using tablet administered NIH toolbox: I-CONECT study

Introduction

Examining the emotional functioning of individuals with mild cognitive impairment (MCI) could help describe their cognitive status and inform the development of interventions. This study compared the emotional characteristics of socially isolated older adults with and without MCI.

Methods

We used baseline data from the Internet-based Conversational Engagement Clinical Trial. Emotional characteristics were assessed with the National Institutes of Health Toolbox Emotion Battery (NIHTB-EB). MCI status was determined with a consensus clinical diagnosis.

Results

This study included 163 participants (mean age = 81.2 years, non-Hispanic Black = 20.7%, MCI = 52.8%). MCI was associated with higher negative affect and lower psychological well-being. Non-Hispanic Black participants scored lower in sadness, higher in positive affect, and higher in meaning and purpose than non-Hispanic White participants.

Conclusion

Older adults with MCI experience more negative emotions and worse psychological well-being than those with normal cognition. The NIHTB-EB appears to be a sensitive tool to detect emotional characteristics associated with cognitive decline.

Can changes in social contact (frequency and mode) mitigate low mood before and during the COVID-19 pandemic? The I-CONECT project

BACKGROUND/OBJECTIVES

The coronavirus disease 2019 (COVID-19) global outbreak allowed a natural experiment to observe how older adults changed social patterns and how it affected their emotional well-being. We studied the frequency and modes of social contact and their effects on older adults' mood before and during the COVID-19 pandemic.

DESIGN

Phone-based surveys were administered weekly before and during the COVID-19 pandemic.

SETTING

Participants were recruited from Portland, Oregon, and Detroit, Michigan.

PARTICIPANTS

Older adults ≥75 years old (n = 155, age = 81.0 ± 4.5, 72.3% women) were included in a randomized controlled trial, the Internet-Based Conversational Engagement Clinical Trial (I-CONECT).

MEASUREMENTS

Low mood was self-reported as feeling downhearted or blue for three or more days in the past week. Social contact was self-reported by the amount of time spent in interactions, with whom (family, friends, others), and via which modes (in-person, phone/video call, text/email/letter).

RESULTS

A total of 5525 weeks of data were derived from 155 participants. Before the COVID-19 pandemic, average social interaction time spent in-person, on phone/video call, and via text/email/letter was 406, 141, and 68 min/week, respectively. During the COVID-19 pandemic, time spent in-person was reduced by 135 min/week, while time spent via phone/video call and writing increased by 33 and 26 mins/week, respectively. In-person family contact was associated with less low mood regardless of the pandemic (odds ratio = 0.92, p < 0.05). There was a COVID-19 × text/email/letter with friends interaction (odds ratio = 0.77, p = 0.03), suggesting that during the COVID-19 pandemic, an increase of 1 h of writing with friends per week was associated with a 23% decrease in the likelihood of experiencing low mood.

CONCLUSION

The lost in-person time relating to COVID-19 restrictions tended to be partially compensated for with increased calls and writing time, although overall social interaction time decreased. During the COVID-19 pandemic, at least two types of social interactions (writing to friends and in-person family time) showed promise for mitigating low mood for older adults with limited social resources.

The Joint Effects of Acoustic and Linguistic Markers for Early Identification of Mild Cognitive Impairment

In recent years, behavioral markers such as spoken language and lexical preferences have been studied in the early detection of mild cognitive impairment (MCI) using conversations. While the combination of linguistic and acoustic signals have been shown to be effective in detecting MCI, they have generally been restricted to structured conversations in which the interviewee responds to fixed prompts. In this study, we show that linguistic and acoustic features can be combined synergistically to identify MCI in semi-structured conversations. Using conversational data from an on-going clinical trial (Clinicaltrials.gov: NCT02871921), we find that the combination of linguistic and acoustic features on semi-structured conversations achieves a mean AUC of 82.7, significantly (p < 0.01) out-performing linguistic-only (74.9 mean AUC) or acoustic-only (65.0 mean AUC) detections on hold-out data. Additionally, features (linguistic, acoustic and combination) obtained from semi-structured conversations outperform their counterparts obtained from structured weekly conversations in identifying MCI. Some linguistic categories are significantly better at predicting MCI status (e.g., death, home) than others.