Skip to content

Intelligent Personal Assistant for Managing Depression in Homebound Older Adults

Intelligent Personal Assistant for Managing Depression in Homebound Older Adults

Status
Completed
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT04900272
Enrollment
21
Registered
2021-05-25
Start date
2020-09-01
Completion date
2022-07-15
Last updated
2024-06-20

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Aging, Physical Dependence

Keywords

Social Isolation, Loneliness, Older Adults, Homebound

Brief summary

The purpose of this study is to design a companion booklet and pilot test it with a voice-controlled intelligent personal assistants (VIPA), like Google Home or Amazon Alexa, to provide homebound older adult patients with skills and tools to help manage social isolation.

Detailed description

The objective of this pilot study is to design a companion booklet and pilot test it with a voice-controlled intelligent personal assistant (VIPA), Google Home or Amazon Alexa, to provide patients with skills and tools to help manage their social isolation. The companion booklet is a document that will be created using feedback obtained from a panel of geriatric experts, highlighting features of the VIPA that may be most beneficial to socially-isolated older adults. It will be submitted to the IRB for approval once finalized. The specific aims of this project are to: Aim 1: Design a companion booklet to be used in conjunction with a VIPA (Google Home/Amazon Alexa) for improving social isolation and communication among homebound older adults, defined as someone who is unable to leave the home without assistance of a device or another person, due to a physical or cognitive condition. Aim 2: Assess the feasibility and implementation of the VIPA (Google Home/Amazon Alexa) and the companion booklet and its impact on clinical and functional outcomes for older adults with social isolation.

Interventions

BEHAVIORALVIPA and companion booklet

Participants will be asked to use the VIPA and accompanying second iteration of the instructional booklet focused on social-isolation. These participants will be surveyed over the phone for 16 weeks (baseline, 2 weeks, 4 weeks, 8 weeks, and 16 weeks) with outcomes measured at each time point.

Sponsors

The Claude D. Pepper Older Americans Independence Centers
CollaboratorOTHER
Northwestern University
Lead SponsorOTHER

Study design

Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
SUPPORTIVE_CARE
Masking
NONE

Eligibility

Sex/Gender
ALL
Age
65 Years to No maximum
Healthy volunteers
Yes

Inclusion criteria

* Adults aged 65 years or older; * Homebound patients of the Northwestern Medicine Geriatrics Program; * English-speaking; and * Able to verbally consent

Exclusion criteria

* Condition that limits ability to consent, such as severe cognitive impairment or lack of English proficiency * Less than 65 years of age

Design outcomes

Primary

MeasureTime frameDescription
Change in Acceptability4 weeks, 8 weeks, and 16 weeksusability assessment that we adapted from this original source: Questionnaire for User Interface Satisfaction. Based on: Lund, A.M. (2001) Measuring Usability with the USE Questionnaire. STC Usability SIG Newsletter, 8:2; constructs assessed: usefulness, ease of use, ease of learning, satisfaction. Responses can range from 1 to 7, and scores represent an average of responses so can also range from 1 to 7, with higher scores representing greater acceptability.
Change in Social Isolationbaseline, 4 weeks, 8 weeks, and 16 weeksPROMIS Social Isolation -Short Form 8a: The PROMIS Social Isolation item bank assesses perceptions of being avoided, excluded, detached, disconnected from, or unknown by, others; scores represent T-scores, with a higher score representing higher levels of social isolation. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 55-60 represents mild social isolation, 60-70 represents moderate, and \>70 represents severe (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).
Change in Anxietybaseline, 4 weeks, 8 weeks, and 16 weeksGeneral Anxiety Disorder Scale (GAD-7): Total score (adding all the numbers) provides a possible score from 0-21, higher scores reflect more anxiety.
Change in Self Efficacy for Managing Social Interactionsbaseline, 4 weeks, 8 weeks, and 16 weeksPROMIS Self Efficacy for Managing Social Interactions -Short Form 4a: : Confidence in participating in social activities and getting help when necessary. Managing communication with others about their medical condition, including communication with health professionals. Scores represent T-scores, with higher scores representing higher self-efficacy. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 30-40 represents low self-efficacy and \<30 represents very low (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).
Change in Companionshipbaseline, 4 weeks, 8 weeks, and 16 weeksPROMIS Companionship -Short Form 4a: The PROMIS adult Companionship item bank assesses perceived availability of someone with whom to share enjoyable social activities such as visiting, talking, celebrations, etc. Scores represent T-scores, with higher scores representing higher levels of companionship. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 30-40 represents low companionship and \<30 represents very low (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).
Change in Ability to Participate in Social Roles and Activitiesbaseline, 4 weeks, 8 weeks, and 16 weeksPROMIS Ability to Participate in Social Roles and Activities -Short Form 4a: Scores represent T-scores, with higher scores representing a higher perceived ability to participate. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 40-45 represents mild difficulty in participating, 30-40 represents moderate, and \<30 represents severe (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).
Change in Depressionbaseline, 4 weeks, 8 weeks, and 16 weeksGeriatric Depression Scale (GDS) - Short Form: scores range from 1 to 15, with higher scores reflecting more depressive symptoms.
Change in Functional Statusbaseline, 4 weeks, 8 weeks, and 16 weeksPROMIS-Physical function 8b: PROMIS Physical Function instruments measure self-reported capability rather than actual performance of physical activities. Scores represent T-scores, with higher scores representing higher levels of physical function. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 40-45 represents mild physical function concerns, 30-40 represents moderate, and \<30 represents severe (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).
Change in Cognitive Functionbaseline, 4 weeks, 8 weeks, and 16 weeksBrief Montreal Cognitive Assessment (MoCA): The Montreal Cognitive Assessment (MoCA) was designed as a rapid screening instrument for mild cognitive dysfunction. It assesses different cognitive domains: attention and concentration, executive functions, memory, language, visuoconstructional skills, conceptual thinking, calculations, and orientation. Scores range from 0 to 22, with higher scores representing higher cognitive function. A score of 18 or higher is considered normal cognition.
Change in Well-beingbaseline, 4 weeks, 8 weeks, and 16 weeksNeuro-QOL Positive Affect & Well-Being: Scores represent T-scores, and higher scores indicate better self-reported well-being. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 30-40 represents low positive affect and \<30 represents very low (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Countries

United States

Participant flow

Participants by arm

ArmCount
VIPA Device and Companion Booklet
For Aim 2, we will recruit a total of 20-40 older adult homebound patients who are likely to be socially isolated, through the Geriatric and General Internal Medicine clinics at Northwestern (over 2,300 homebound older adults) to use the VIPA and accompanying instructional booklet focused on social-isolation. Either the study PI or Research Study Coordinator (RSC) will visit the participant's home, or assist the patient virtually, once consented, to set up the Google Home or Amazon Alexa device for them and review the companion VIPA booklet with them. VIPA and companion booklet: Participants will be asked to use the VIPA and accompanying second iteration of the instructional booklet focused on social-isolation. These participants will be surveyed over the phone for 16 weeks (baseline, 2 weeks, 4 weeks, 8 weeks, and 16 weeks) with outcomes measured at each time point.
17
Total17

Withdrawals & dropouts

PeriodReasonFG000
Overall StudyLost to Follow-up2
Overall StudyWithdrawal by Subject1
Overall StudyWithdrawal due to technical difficulties (see above)3

Baseline characteristics

CharacteristicVIPA Device and Companion Booklet
Age, Continuous80.53 years
STANDARD_DEVIATION 8.12
Ethnicity (NIH/OMB)
Hispanic or Latino
0 Participants
Ethnicity (NIH/OMB)
Not Hispanic or Latino
17 Participants
Ethnicity (NIH/OMB)
Unknown or Not Reported
0 Participants
Race (NIH/OMB)
American Indian or Alaska Native
0 Participants
Race (NIH/OMB)
Asian
1 Participants
Race (NIH/OMB)
Black or African American
4 Participants
Race (NIH/OMB)
More than one race
0 Participants
Race (NIH/OMB)
Native Hawaiian or Other Pacific Islander
0 Participants
Race (NIH/OMB)
Unknown or Not Reported
1 Participants
Race (NIH/OMB)
White
11 Participants
Region of Enrollment
United States
17 participants
Sex: Female, Male
Female
12 Participants
Sex: Female, Male
Male
5 Participants

Adverse events

Event typeEG000
affected / at risk
deaths
Total, all-cause mortality
0 / 17
other
Total, other adverse events
0 / 17
serious
Total, serious adverse events
0 / 17

Outcome results

Primary

Change in Ability to Participate in Social Roles and Activities

PROMIS Ability to Participate in Social Roles and Activities -Short Form 4a: Scores represent T-scores, with higher scores representing a higher perceived ability to participate. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 40-45 represents mild difficulty in participating, 30-40 represents moderate, and \<30 represents severe (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Ability to Participate in Social Roles and ActivitiesBaseline52.97 T-scoreStandard Deviation 6.72
VIPA Device and Companion BookletChange in Ability to Participate in Social Roles and Activities4 weeks50.6 T-scoreStandard Deviation 8.42
VIPA Device and Companion BookletChange in Ability to Participate in Social Roles and Activities8 weeks50.39 T-scoreStandard Deviation 8.23
VIPA Device and Companion BookletChange in Ability to Participate in Social Roles and Activities16 weeks50.04 T-scoreStandard Deviation 7.99
p-value: 0.728ANOVA
Primary

Change in Acceptability

usability assessment that we adapted from this original source: Questionnaire for User Interface Satisfaction. Based on: Lund, A.M. (2001) Measuring Usability with the USE Questionnaire. STC Usability SIG Newsletter, 8:2; constructs assessed: usefulness, ease of use, ease of learning, satisfaction. Responses can range from 1 to 7, and scores represent an average of responses so can also range from 1 to 7, with higher scores representing greater acceptability.

Time frame: 4 weeks, 8 weeks, and 16 weeks

Population: one participant had to miss each of the last two timepoints for logistical reasons

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Acceptability4 weeks5.2 score on a scaleStandard Deviation 1.36
VIPA Device and Companion BookletChange in Acceptability8 weeks5.27 score on a scaleStandard Deviation 1.44
VIPA Device and Companion BookletChange in Acceptability16 weeks5.34 score on a scaleStandard Deviation 1.53
p-value: 0.785ANOVA
Primary

Change in Anxiety

General Anxiety Disorder Scale (GAD-7): Total score (adding all the numbers) provides a possible score from 0-21, higher scores reflect more anxiety.

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in AnxietyBaseline5.44 score on a scaleStandard Deviation 5.85
VIPA Device and Companion BookletChange in Anxiety4 weeks3.87 score on a scaleStandard Deviation 4
VIPA Device and Companion BookletChange in Anxiety8 weeks5.57 score on a scaleStandard Deviation 4.82
VIPA Device and Companion BookletChange in Anxiety16 weeks3.86 score on a scaleStandard Deviation 4.52
p-value: 0.5015ANOVA
Primary

Change in Cognitive Function

Brief Montreal Cognitive Assessment (MoCA): The Montreal Cognitive Assessment (MoCA) was designed as a rapid screening instrument for mild cognitive dysfunction. It assesses different cognitive domains: attention and concentration, executive functions, memory, language, visuoconstructional skills, conceptual thinking, calculations, and orientation. Scores range from 0 to 22, with higher scores representing higher cognitive function. A score of 18 or higher is considered normal cognition.

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: after baseline, 1 participant withdrew, and 1 participant missed each of the following timepoints. Only 7 participants completed this measure due to technical difficulties and hearing impairments that impeded participants' ability to complete this measure virtually.

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Cognitive FunctionBaseline15.71 score on a scaleStandard Deviation 4.31
VIPA Device and Companion BookletChange in Cognitive Function4 weeks15 score on a scaleStandard Deviation 3.16
VIPA Device and Companion BookletChange in Cognitive Function8 weeks16 score on a scaleStandard Deviation 4.69
VIPA Device and Companion BookletChange in Cognitive Function16 weeks16 score on a scaleStandard Deviation 4.04
p-value: 0.3356ANOVA
Primary

Change in Companionship

PROMIS Companionship -Short Form 4a: The PROMIS adult Companionship item bank assesses perceived availability of someone with whom to share enjoyable social activities such as visiting, talking, celebrations, etc. Scores represent T-scores, with higher scores representing higher levels of companionship. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 30-40 represents low companionship and \<30 represents very low (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in CompanionshipBaseline48.26 T-scoreStandard Deviation 7.27
VIPA Device and Companion BookletChange in Companionship4 weeks45.06 T-scoreStandard Deviation 9.03
VIPA Device and Companion BookletChange in Companionship8 weeks48.72 T-scoreStandard Deviation 7.99
VIPA Device and Companion BookletChange in Companionship16 weeks46.36 T-scoreStandard Deviation 10.35
p-value: 0.2688ANOVA
Primary

Change in Depression

Geriatric Depression Scale (GDS) - Short Form: scores range from 1 to 15, with higher scores reflecting more depressive symptoms.

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in DepressionBaseline3.31 score on a scaleStandard Deviation 2.65
VIPA Device and Companion BookletChange in Depression4 weeks3.8 score on a scaleStandard Deviation 3.12
VIPA Device and Companion BookletChange in Depression8 weeks2.54 score on a scaleStandard Deviation 1.71
VIPA Device and Companion BookletChange in Depression16 weeks3.45 score on a scaleStandard Deviation 2.7
p-value: 0.473ANOVA
Primary

Change in Functional Status

PROMIS-Physical function 8b: PROMIS Physical Function instruments measure self-reported capability rather than actual performance of physical activities. Scores represent T-scores, with higher scores representing higher levels of physical function. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 40-45 represents mild physical function concerns, 30-40 represents moderate, and \<30 represents severe (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Functional StatusBaseline33.46 T-scoreStandard Deviation 7.82
VIPA Device and Companion BookletChange in Functional Status4 weeks33.18 T-scoreStandard Deviation 6.82
VIPA Device and Companion BookletChange in Functional Status8 weeks31.7 T-scoreStandard Deviation 6.93
VIPA Device and Companion BookletChange in Functional Status16 weeks31.4 T-scoreStandard Deviation 7.99
p-value: 0.4161ANOVA
Primary

Change in Self Efficacy for Managing Social Interactions

PROMIS Self Efficacy for Managing Social Interactions -Short Form 4a: : Confidence in participating in social activities and getting help when necessary. Managing communication with others about their medical condition, including communication with health professionals. Scores represent T-scores, with higher scores representing higher self-efficacy. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 30-40 represents low self-efficacy and \<30 represents very low (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Self Efficacy for Managing Social InteractionsBaseline47.59 T-scoreStandard Deviation 8.05
VIPA Device and Companion BookletChange in Self Efficacy for Managing Social Interactions4 weeks48.54 T-scoreStandard Deviation 9.19
VIPA Device and Companion BookletChange in Self Efficacy for Managing Social Interactions8 weeks49.4 T-scoreStandard Deviation 9.18
VIPA Device and Companion BookletChange in Self Efficacy for Managing Social Interactions16 weeks45.96 T-scoreStandard Deviation 8.15
p-value: 0.3536ANOVA
Primary

Change in Social Isolation

PROMIS Social Isolation -Short Form 8a: The PROMIS Social Isolation item bank assesses perceptions of being avoided, excluded, detached, disconnected from, or unknown by, others; scores represent T-scores, with a higher score representing higher levels of social isolation. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 55-60 represents mild social isolation, 60-70 represents moderate, and \>70 represents severe (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Social Isolation16 weeks47.75 T-scoreStandard Deviation 10.93
VIPA Device and Companion BookletChange in Social IsolationBaseline48.04 T-scoreStandard Deviation 8.54
VIPA Device and Companion BookletChange in Social Isolation4 weeks49.44 T-scoreStandard Deviation 8.17
VIPA Device and Companion BookletChange in Social Isolation8 weeks47.44 T-scoreStandard Deviation 9.4
p-value: 0.583ANOVA
Primary

Change in Well-being

Neuro-QOL Positive Affect & Well-Being: Scores represent T-scores, and higher scores indicate better self-reported well-being. T-score of 50 indicates the population mean, with a standard deviation of 10. In terms of cut points, a T-score of 30-40 represents low positive affect and \<30 represents very low (https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points).

Time frame: baseline, 4 weeks, 8 weeks, and 16 weeks

Population: one participant withdrew after baseline, and one participant missed each of the final two timepoints

ArmMeasureGroupValue (MEAN)Dispersion
VIPA Device and Companion BookletChange in Well-beingBaseline54.27 T-scoreStandard Deviation 7.25
VIPA Device and Companion BookletChange in Well-being4 weeks54.73 T-scoreStandard Deviation 7.08
VIPA Device and Companion BookletChange in Well-being8 weeks55.21 T-scoreStandard Deviation 9.32
VIPA Device and Companion BookletChange in Well-being16 weeks54.66 T-scoreStandard Deviation 9
p-value: 0.9825ANOVA

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026