The Creyos online cognitive test battery included digit span, feature match, grammatical reasoning, paired associates, and token search. We converted raw scores to standardized z scores for each test using the group baseline mean and standard deviation. We then summed the standardized scores to create a composite cognitive score and re-standardized the composite score using the group baseline mean and standard deviation (baseline mean=0, SD=1). Positive change reflects improved cognitive performance.

The Brain Health Together and Brain Health Together Plus group interventions were identical for the first 3 months. Therefore, we decided a priori to combine them for all 3-month analyses to maximize statistical power.

This is a validated measure that combines validated component measures of individual dementia risk factors and also provides a summary dementia risk score. We created a modified version that only included risk factors that are modifiable and targeted by our intervention (depression, physical activity, cognitive activity, social isolation, alcohol consumption, smoking, and fish intake) using the weights from the original score. We then converted the calculated score to a z score using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased dementia risk; negative scores reflect reduced dementia risk.

The Brain Health Together and Brain Health Together Plus groups received identical interventions for the first 3 months. Therefore, we decided a priori to combine these groups for all 3-month analyses to maximize statistical power.

The Creyos online cognitive test battery included digit span, feature match, grammatical reasoning, paired associates, and token search. We converted raw scores to standardized z scores for each test using the group baseline mean and standard deviation. We then summed the standardized scores to create a composite cognitive score and re-standardized the composite score using the group baseline mean and standard deviation (baseline mean=0, SD=1). Positive change reflects improved cognitive performance.

This is a validated measure that combines validated component measures of individual dementia risk factors and also provides a summary dementia risk score. We created a modified version that only included risk factors that are modifiable and targeted by our intervention (depression, physical activity, cognitive activity, social isolation, alcohol consumption, smoking, and fish intake) using the weights from the original score. We then converted the calculated score to a z-score using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased dementia risk; negative scores reflect reduced dementia risk.

Loneliness / social isolation was measured using the validated 3-item UCLA Loneliness Scale, which consists of 3 items (lack companionship, feel left out, feel isolated) that are rated as 1 (hardly ever), 2 (some of the time), or 3 (often). Scores may range from 3 to 9, with higher scores reflecting greater loneliness. Raw scores were converted to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased loneliness, negative change scores reflect decreased loneliness.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We assessed depressive symptoms using the validated 10-item version of the Center for Epidemiologic Studies-Depression Scale (CES-D). Participants are asked about feelings in the past week with response categories of 0 (<1/day), 1 (1-2 days), 2 (3-4 days) or 3 (5-7 days). Scores may range from 0 to 30, with higher scores indicating more depressive symptoms. We concerted raw scores to z-scores using the baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased depressive symptoms; negative change scores reflect reduced depressive symptoms.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We measured engagement in cognitive stimulating activities using the validated Cognitive Activity Questionnaire (CAQ). The CAQ includes 11 items that ask how often individuals engage in various cognitive activities such as reading or playing games on a 6-point Likert scale (once a month/never to every day). Higher scores reflect more cognitive activity. We converted raw scores to standardized z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased cognitive activity; negative change scores reflect reduced cognitive activity.

We decided a priori to combine Brain Health Together and Brain Health Together groups for all 3-month analyses.

Physical activity was assessed using the using the validated International Physical Activity Questionnaire short form. Participants were asked about time spent (days/week, minutes/day) doing vigorous physical activity, moderate physical activity, walking and sitting in the past 7 days, and responses were converted into metabolic equivalents per week. We converted raw scores into z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased physical activity; negative change scores reflect reduced physical activity.

The Brain Health Together and Brain Health Together Plus groups received identical interventions for the first 3 months. Therefore, we decided a priori to combine these groups for all 3-month analyses to maximize statistical power.

We assessed emotional well-being using the Neuro-QOL v1.0 Positive Affect and Well-Being Short Form, which includes 9 items (e.g. sense of well-being, feeling hopeful, life was satisfying, etc.) with 5-point responses from never (1) to always (5). Scores may range from 9 to 45, with higher scores reflecting greater feelings of well-being. We converted raw scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased well-being; negative change scores reflect decreased emotional well-being.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We measured mind-body awareness using the self-regulation sub-scale of the validated Multidimensional Assessment of Interoceptive Awareness version 2 (MAIA-2). The self-regulation sub-scale includes 4 items rated on a six-point Likert scale (0/never to 5/always) about the ability to regulate distress by paying attention to bodily sensations (e.g., "When I feel overwhelmed I can find a calm place inside.") Scores reflect the mean of responses (range 0 to 5), with higher scores indicating greater self-regulation. We converted raw score to z-scores using the baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved mind-body awareness; negative change scores reflect worse mind-body awareness.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We measured diet using the validated MIND diet score. The MIND diet is a hybrid of the Mediterranean and DASH (Dietary Approaches to Stop Hypertension) diets. It has 15 dietary components including 10 brain healthy foods (e.g., green leafy vegetables, berries) and 5 less healthy foods (e.g., red meat, fried/fast foods). The frequency of consumption for each item is scored as 0, 0.5 or 1, and the total MIND diet score is calculated by summing over the 15 components. Scores may range from 0 to 15, with higher scores reflecting better adherence to the MIND diet. We converted raw scores to z-scores using the baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved diet; negative change scores reflect worse diet.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We measured sleep disturbance using the validated PROMIS Sleep Disturbance Short Form 4-item measure, which asks about sleep quality and disturbance in the past 7 days on a 5-point Likert scale. Raw scores are converted to t-scores. Higher scores indicate more sleep disturbance. We converted t-scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased sleep disturbance; negative change scores reflect reduced sleep disturbance.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We used a modified version of the Self-Efficacy for Managing Chronic Disease 6-item scale to measure self-efficacy for engaging in activities to support brain health. Each item is rated on an 11-point Likert scale from 0 (not at all confident) to 10 (totally confident). The score is the mean of the 6 items with higher scores reflecting higher self-efficacy. We converted raw scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved self-efficacy; negative change scores reflect worse self-efficacy.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We measured global quality of life using the validated PROMIS Scale v1.2 - Global Health. This scale includes 10 items rated on a 5-point Likert scale from 1 (poor) to 5 (excellent) and includes questions about overall self-rated health, overall quality of life, physical health, mental health, social relationships, fatigue and pain. Raw scores are converted into T-scores for physical and mental quality of life. We converted t-scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved quality of life; negative change scores reflect worsened quality of life.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

We measured global quality of life using the validated PROMIS Scale v1.2 - Global Health. This scale includes 10 items rated on a 5-point Likert scale from 1 (poor) to 5 (excellent) and includes questions about overall self-rated health, overall quality of life, physical health, mental health, social relationships, fatigue and pain. Raw scores are converted into T-scores for physical and mental quality of life. We converted t-scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved quality of life; negative change scores reflect worsened quality of life.

We decided a priori to combine the Brain Health Together and Brain Health Together Plus groups for all 3-month analyses.

Loneliness / social isolation was measured using the validated 3-item UCLA Loneliness Scale, which consists of 3 items (lack companionship, feel left out, feel isolated) that are rated as 1 (hardly ever), 2 (some of the time), or 3 (often). Scores may range from 3 to 9, with higher scores reflecting greater loneliness. Raw scores were converted to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased loneliness, negative change scores reflect decreased loneliness.

We assessed depressive symptoms using the validated 10-item version of the Center for Epidemiologic Studies-Depression Scale (CES-D). Participants are asked about feelings in the past week with response categories of 0 (<1/day), 1 (1-2 days), 2 (3-4 days) or 3 (5-7 days). Scores may range from 0 to 30, with higher scores indicating more depressive symptoms. We concerted raw scores to z-scores using the baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased depressive symptoms; negative change scores reflect reduced depressive symptoms.

We measured engagement in cognitive stimulating activities using the validated Cognitive Activity Questionnaire (CAQ). The CAQ includes 11 items that ask how often individuals engage in various cognitive activities such as reading or playing games on a 6-point Likert scale (once a month/never to every day). Higher scores reflect more cognitive activity. We converted raw scores to standardized z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased cognitive activity; negative change scores reflect reduced cognitive activity.

Physical activity was assessed using the using the validated International Physical Activity Questionnaire short form. Participants were asked about time spent (days/week, minutes/day) doing vigorous physical activity, moderate physical activity, walking and sitting in the past 7 days, and responses were converted into metabolic equivalents per week. We converted raw scores into z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased physical activity; negative change scores reflect reduced physical activity.

We assessed emotional well-being using the Neuro-QOL v1.0 Positive Affect and Well-Being Short Form, which includes 9 items (e.g. sense of well-being, feeling hopeful, life was satisfying, etc.) with 5-point responses from never (1) to always (5). Scores may range from 9 to 45, with higher scores reflecting greater feelings of well-being. We converted raw scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased well-being; negative change scores reflect decreased emotional well-being.

We measured mind-body awareness using the self-regulation sub-scale of the validated Multidimensional Assessment of Interoceptive Awareness version 2 (MAIA-2). The self-regulation sub-scale includes 4 items rated on a six-point Likert scale (0/never to 5/always) about the ability to regulate distress by paying attention to bodily sensations (e.g., "When I feel overwhelmed I can find a calm place inside.") Scores reflect the mean of responses (range 0 to 5), with higher scores indicating greater self-regulation. We converted raw score to z-scores using the baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved mind-body awareness; negative change scores reflect worse mind-body awareness.

We measured diet using the validated MIND diet score. The MIND diet is a hybrid of the Mediterranean and DASH (Dietary Approaches to Stop Hypertension) diets. It has 15 dietary components including 10 brain healthy foods (e.g., green leafy vegetables, berries) and 5 less healthy foods (e.g., red meat, fried/fast foods). The frequency of consumption for each item is scored as 0, 0.5 or 1, and the total MIND diet score is calculated by summing over the 15 components. Scores may range from 0 to 15, with higher scores reflecting better adherence to the MIND diet. We converted raw scores to z-scores using the baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved diet; negative change scores reflect worse diet.

We measured sleep disturbance using the validated PROMIS Sleep Disturbance Short Form 4-item measure, which asks about sleep quality and disturbance in the past 7 days on a 5-point Likert scale. Raw scores are converted to t-scores. Higher scores indicate more sleep disturbance. We converted t-scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect increased sleep disturbance; negative change scores reflect reduced sleep disturbance.

We used a modified version of the Self-Efficacy for Managing Chronic Disease 6-item scale to measure self-efficacy for engaging in activities to support brain health. Each item is rated on an 11-point Likert scale from 0 (not at all confident) to 10 (totally confident). The score is the mean of the 6 items with higher scores reflecting higher self-efficacy. We converted raw scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved self-efficacy; negative change scores reflect worse self-efficacy.

We measured global quality of life using the validated PROMIS Scale v1.2 - Global Health. This scale includes 10 items rated on a 5-point Likert scale from 1 (poor) to 5 (excellent) and includes questions about overall self-rated health, overall quality of life, physical health, mental health, social relationships, fatigue and pain. Raw scores are converted into T-scores for physical and mental quality of life. We converted t-scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved quality of life; negative change scores reflect worsened quality of life.

We measured global quality of life using the validated PROMIS Scale v1.2 - Global Health. This scale includes 10 items rated on a 5-point Likert scale from 1 (poor) to 5 (excellent) and includes questions about overall self-rated health, overall quality of life, physical health, mental health, social relationships, fatigue and pain. Raw scores are converted into T-scores for physical and mental quality of life. We converted t-scores to z-scores using the group baseline mean and SD (baseline mean=0, SD=1). Positive change scores reflect improved quality of life; negative change scores reflect worsened quality of life.