Trial Outcomes & Findings for Digital Detection of Dementia (D Cubed) Studies: D2 (NCT NCT05231954)
NCT ID: NCT05231954
Last Updated: 2025-08-20
Results Overview
Any new ADRD case identified (documented in the EHR) within 12 months of the Annual Wellness Visit (index visit).
Recruitment status
COMPLETED
Study phase
NA
Target enrollment
5325 participants
Primary outcome timeframe
12 months after index visit
Results posted on
2025-08-20
Participant Flow
D2 trial was launched at 9 clinics on July 5, 2022 (3 clinics in each study arm). D2 trial was completed in July 2024. The D2 trial deployed the Passive Digital Marker (PDM) on 5,325 older patients who were eligible and enrolled to receive primary care services within one of the 9 randomized clinics.
* At least 65 years old as of date of data pull. * At least 3 years of EHR data as of date of data pull. * Alive as of date of data pull. * No prior ADRD, MCI, bipolar, or schizophrenia diagnosis as determined by ICD-10 code. * No evidence for any history of cholinesterase inhibitors or memantine prescription." Within each clinic, enrolled patients include only those who: * Meet the above eligibility criteria * Have at least one visit to a randomized clinic during the study period.
Unit of analysis: Clinics
Participant milestones
| Measure |
Annual Well Visit or Any Other Visit to Primary Care Doctor
Annual Well Visit or any other visit to Primary Care Doctor: This is the usual care arm. Electronic Health Record Data for patients from the clinics randomized to usual care will be collected for comparison with the other 2 arms. Patients from these primary care clinics must have had a visit to their doctor either as an annual well visit (AWV) or any other type of visit. These clinics will not have to do anything for the study but run their business as usual without altering anything.
|
Passive Digital Marker (PDM)
Passive Digital Marker (PDM): Electronic Health Record Data from those clinics randomized to PDM will be run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset.
|
Passive Digital Marker (PDM) + Quick Dementia Rating Scale (QDRS)
Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
|
|---|---|---|---|
|
Overall Study
STARTED
|
1724 3
|
1300 3
|
2301 3
|
|
Overall Study
COMPLETED
|
1724 3
|
1300 3
|
2301 3
|
|
Overall Study
NOT COMPLETED
|
0 0
|
0 0
|
0 0
|
Reasons for withdrawal
Withdrawal data not reported
Baseline Characteristics
Digital Detection of Dementia (D Cubed) Studies: D2
Baseline characteristics by cohort
| Measure |
Annual Well Visit or Any Other Visit to Primary Care Doctor
n=1724 Participants
Annual Well Visit or any other visit to Primary Care Doctor: This is the usual care arm. Electronic Health Record Data for patients from the clinics randomized to usual care will be collected for comparison with the other 2 arms. Patients from these primary care clinics must have had a visit to their doctor either as an annual well visit (AWV) or any other type of visit. These clinics will not have to do anything for the study but run their business as usual without altering anything.
|
Passive Digital Marker (PDM)
n=1300 Participants
Passive Digital Marker (PDM): Electronic Health Record Data from those clinics randomized to PDM will be run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset.
|
Passive Digital Marker (PDM) + Quick Dementia Rating Scale (QDRS)
n=2301 Participants
Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
|
Total
n=5325 Participants
Total of all reporting groups
|
|---|---|---|---|---|
|
Age, Continuous
|
71 years
STANDARD_DEVIATION 6 • n=99 Participants
|
71 years
STANDARD_DEVIATION 6 • n=107 Participants
|
71 years
STANDARD_DEVIATION 6 • n=206 Participants
|
71 years
STANDARD_DEVIATION 6 • n=7 Participants
|
|
Sex/Gender, Customized
Sex/Gender · Female
|
1036 Participants
n=99 Participants
|
827 Participants
n=107 Participants
|
1449 Participants
n=206 Participants
|
3312 Participants
n=7 Participants
|
|
Sex/Gender, Customized
Sex/Gender · Male
|
688 Participants
n=99 Participants
|
473 Participants
n=107 Participants
|
852 Participants
n=206 Participants
|
2013 Participants
n=7 Participants
|
|
Race (NIH/OMB)
American Indian or Alaska Native
|
9 Participants
n=99 Participants
|
12 Participants
n=107 Participants
|
9 Participants
n=206 Participants
|
30 Participants
n=7 Participants
|
|
Race (NIH/OMB)
Asian
|
10 Participants
n=99 Participants
|
13 Participants
n=107 Participants
|
14 Participants
n=206 Participants
|
37 Participants
n=7 Participants
|
|
Race (NIH/OMB)
Native Hawaiian or Other Pacific Islander
|
0 Participants
n=99 Participants
|
0 Participants
n=107 Participants
|
0 Participants
n=206 Participants
|
0 Participants
n=7 Participants
|
|
Race (NIH/OMB)
Black or African American
|
986 Participants
n=99 Participants
|
605 Participants
n=107 Participants
|
1336 Participants
n=206 Participants
|
2927 Participants
n=7 Participants
|
|
Race (NIH/OMB)
White
|
575 Participants
n=99 Participants
|
471 Participants
n=107 Participants
|
743 Participants
n=206 Participants
|
1789 Participants
n=7 Participants
|
|
Race (NIH/OMB)
More than one race
|
104 Participants
n=99 Participants
|
157 Participants
n=107 Participants
|
157 Participants
n=206 Participants
|
418 Participants
n=7 Participants
|
|
Race (NIH/OMB)
Unknown or Not Reported
|
40 Participants
n=99 Participants
|
42 Participants
n=107 Participants
|
42 Participants
n=206 Participants
|
124 Participants
n=7 Participants
|
PRIMARY outcome
Timeframe: 12 months after index visitPopulation: 9 clinics were randomized, and patients were treated as per the randomization arm (3 clinics per arm)
Any new ADRD case identified (documented in the EHR) within 12 months of the Annual Wellness Visit (index visit).
Outcome measures
| Measure |
Annual Well Visit or Any Other Visit to Primary Care Doctor
n=1724 Participants
Annual Well Visit or any other visit to Primary Care Doctor: This is the usual care arm. Electronic Health Record Data for patients from the clinics randomized to usual care will be collected for comparison with the other 2 arms. Patients from these primary care clinics must have had a visit to their doctor either as an annual well visit (AWV) or any other type of visit. These clinics will not have to do anything for the study but run their business as usual without altering anything.
|
Passive Digital Marker (PDM)
n=1300 Participants
Passive Digital Marker (PDM): Electronic Health Record Data from those clinics randomized to PDM will be run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset.
|
Passive Digital Marker (PDM) + Quick Dementia Rating Scale (QDRS)
n=2301 Participants
Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
|
|---|---|---|---|
|
12-Month Cumulative Incidence of ADRD Diagnoses
|
213 Number of new ADRD cases
|
134 Number of new ADRD cases
|
355 Number of new ADRD cases
|
SECONDARY outcome
Timeframe: 12 months after index visitPopulation: 9 clinics were randomized, and patients were treated as per the randomization arm (3 clinics per arm)
The secondary outcome measures is receipt of any services related to cognitive diagnostic assessment in the post Annual Wellness Visit (index) period that providers may order to diagnose or exclude ADRD. Specifically, the metrics of diagnostic assessment are evaluated as proportions of patients with a record of 1 or more of: * Laboratory tests for TSH, serum B12, folate, or syphilis; individually or combined at any point during the 90 days after index * Neuropsychological testing, including testing by psychologist or physician, technician administrator, computer, or other providers during the 12 months after index date * Brain imaging testing (computed tomography, magnetic resonance imaging, positron emission tomography, magnetic resonance angiogram) of the head and neck, brain, or skull during the 12 months after index date * Medications approved for management of ADRD (cholinesterase inhibitors, memantine) during the 12 months after index date
Outcome measures
| Measure |
Annual Well Visit or Any Other Visit to Primary Care Doctor
n=1724 Participants
Annual Well Visit or any other visit to Primary Care Doctor: This is the usual care arm. Electronic Health Record Data for patients from the clinics randomized to usual care will be collected for comparison with the other 2 arms. Patients from these primary care clinics must have had a visit to their doctor either as an annual well visit (AWV) or any other type of visit. These clinics will not have to do anything for the study but run their business as usual without altering anything.
|
Passive Digital Marker (PDM)
n=1300 Participants
Passive Digital Marker (PDM): Electronic Health Record Data from those clinics randomized to PDM will be run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset.
|
Passive Digital Marker (PDM) + Quick Dementia Rating Scale (QDRS)
n=2301 Participants
Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
|
|---|---|---|---|
|
12-Month Cumulative Incidence of ADRD Services
|
879 Participants with ADRD related servicess
|
636 Participants with ADRD related servicess
|
1366 Participants with ADRD related servicess
|
Adverse Events
Annual Well Visit or Any Other Visit to Primary Care Doctor
Serious events: 561 serious events
Other events: 0 other events
Deaths: 26 deaths
Passive Digital Marker (PDM)
Serious events: 398 serious events
Other events: 0 other events
Deaths: 7 deaths
Passive Digital Marker (PDM) + Quick Dementia Rating Scale (QDRS)
Serious events: 816 serious events
Other events: 0 other events
Deaths: 25 deaths
Serious adverse events
| Measure |
Annual Well Visit or Any Other Visit to Primary Care Doctor
n=1724 participants at risk
Annual Well Visit or any other visit to Primary Care Doctor: This is the usual care arm. Electronic Health Record Data for patients from the clinics randomized to usual care will be collected for comparison with the other 2 arms. Patients from these primary care clinics must have had a visit to their doctor either as an annual well visit (AWV) or any other type of visit. These clinics will not have to do anything for the study but run their business as usual without altering anything.
|
Passive Digital Marker (PDM)
n=1300 participants at risk
Passive Digital Marker (PDM): Electronic Health Record Data from those clinics randomized to PDM will be run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset.
Passive Digital Marker for screening for ADRD: Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
|
Passive Digital Marker (PDM) + Quick Dementia Rating Scale (QDRS)
n=2301 participants at risk
Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
Passive Digital Marker for screening for ADRD: Patients in the primary care clinics randomized to PDM+QDRS will have Electronic Health Record Data of their patients run through the PDM, a machine learning algorithm which can predict ADRD one year and three years prior to its onset. In addition, patients from these clinics will have their patients complete the QDRS, a validated patient reported outcome (PRO) tool. This combined approach will assess the value of early detection of ADRD and if the annual well visit can overcome the barriers related to early detection of ADRD.
|
|---|---|---|---|
|
General disorders
Positive for PDM and had Hospitalization
|
15.7%
88/561 • Number of events 88 • 12-Months of follow-up after index visit (enrollment date)
|
16.3%
65/398 • Number of events 65 • 12-Months of follow-up after index visit (enrollment date)
|
18.1%
148/816 • Number of events 148 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had ER visit
|
43.3%
243/561 • Number of events 243 • 12-Months of follow-up after index visit (enrollment date)
|
44.0%
175/398 • Number of events 175 • 12-Months of follow-up after index visit (enrollment date)
|
48.0%
392/816 • Number of events 392 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had PCC visit
|
100.0%
561/561 • Number of events 561 • 12-Months of follow-up after index visit (enrollment date)
|
100.0%
398/398 • Number of events 398 • 12-Months of follow-up after index visit (enrollment date)
|
100.0%
816/816 • Number of events 816 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had Memory clinic visit
|
5.9%
33/561 • Number of events 33 • 12-Months of follow-up after index visit (enrollment date)
|
8.3%
33/398 • Number of events 33 • 12-Months of follow-up after index visit (enrollment date)
|
7.6%
62/816 • Number of events 62 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had NEW Dementia Diagnosis
|
15.5%
87/561 • Number of events 87 • 12-Months of follow-up after index visit (enrollment date)
|
17.6%
70/398 • Number of events 70 • 12-Months of follow-up after index visit (enrollment date)
|
22.8%
186/816 • Number of events 186 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had NEW Depression Diagnosis
|
5.5%
31/561 • Number of events 31 • 12-Months of follow-up after index visit (enrollment date)
|
4.5%
18/398 • Number of events 18 • 12-Months of follow-up after index visit (enrollment date)
|
3.9%
32/816 • Number of events 32 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had NEW Suicidal Ideation Diagnosis
|
0.36%
2/561 • Number of events 2 • 12-Months of follow-up after index visit (enrollment date)
|
0.25%
1/398 • Number of events 1 • 12-Months of follow-up after index visit (enrollment date)
|
0.37%
3/816 • Number of events 3 • 12-Months of follow-up after index visit (enrollment date)
|
|
General disorders
Positive for PDM and had NEW significant incapacity Diagnosis
|
6.2%
35/561 • Number of events 35 • 12-Months of follow-up after index visit (enrollment date)
|
7.3%
29/398 • Number of events 29 • 12-Months of follow-up after index visit (enrollment date)
|
5.0%
41/816 • Number of events 41 • 12-Months of follow-up after index visit (enrollment date)
|
Other adverse events
Adverse event data not reported
Additional Information
Results disclosure agreements
- Principal investigator is a sponsor employee
- Publication restrictions are in place