Trial Outcomes & Findings for Data-driven Identification for Substance Misuse (NCT NCT03833804)
NCT ID: NCT03833804
Last Updated: 2025-10-24
Results Overview
The primary outcome is the proportion of patients who received SBIRT after a positive universal screen for being at risk for substance misuse. The design is an interrupted time-series prospective observational study.
Recruitment status
COMPLETED
Study phase
NA
Target enrollment
64996 participants
Primary outcome timeframe
24 months
Results posted on
2025-10-24
Participant Flow
The comparison groups are pre-intervention and intervention. data from 31432 people were assessed prior to the intervention in 2022-2023 and data from 33564 people were assessed with the intervention from 2023-2024: data from 64996 people were assessed in total.
Participant milestones
| Measure |
Usual Care
Before intervention
|
NLP (Natural Language Processing) Pre-screen
Automated processing of clinical notes collected during routine care in first 24 hours of hospital admission to identify individuals at-risk for substance misuse to receive standard-of-care full screening and assessment, brief intervention, or referral to treatment (SBIRT) intervention.
Processing of clinical notes in the EHR data collected during routine care: Clinical notes collected in the first day of hospital admission during usual care as input to natural language processing and machine learning algorithm.
|
|---|---|---|
|
Usual Care Data Collection (2022-2023)
STARTED
|
31432
|
0
|
|
Usual Care Data Collection (2022-2023)
COMPLETED
|
31432
|
0
|
|
Usual Care Data Collection (2022-2023)
NOT COMPLETED
|
0
|
0
|
|
SMART-AI Data Collection (2023-2024)
STARTED
|
0
|
33564
|
|
SMART-AI Data Collection (2023-2024)
COMPLETED
|
0
|
33564
|
|
SMART-AI Data Collection (2023-2024)
NOT COMPLETED
|
0
|
0
|
Reasons for withdrawal
Withdrawal data not reported
Baseline Characteristics
Elixhauser Comorbidity was unknown for some participants.
Baseline characteristics by cohort
| Measure |
Usual Care
n=31432 Participants
Before intervention
|
NLP (Natural Language Processing) Pre-screen: SMART-AI
n=33564 Participants
Automated processing of clinical notes collected during routine care in first 24 hours of hospital admission to identify individuals at-risk for substance misuse to receive standard-of-care full screening and assessment, brief intervention, or referral to treatment (SBIRT) intervention.
Processing of clinical notes in the EHR data collected during routine care: Clinical notes collected in the first day of hospital admission during usual care as input to natural language processing and machine learning algorithm.
|
Total
n=64996 Participants
Total of all reporting groups
|
|---|---|---|---|
|
Age, Continuous
|
56 years
STANDARD_DEVIATION 19 • n=31432 Participants
|
56 years
STANDARD_DEVIATION 19 • n=33564 Participants
|
56 years
STANDARD_DEVIATION 19 • n=64996 Participants
|
|
Sex: Female, Male
Female
|
18458 Participants
n=31432 Participants
|
19421 Participants
n=33564 Participants
|
37879 Participants
n=64996 Participants
|
|
Sex: Female, Male
Male
|
12974 Participants
n=31432 Participants
|
14143 Participants
n=33564 Participants
|
27117 Participants
n=64996 Participants
|
|
Race/Ethnicity, Customized
White
|
10524 Participants
n=31432 Participants
|
11097 Participants
n=33564 Participants
|
21621 Participants
n=64996 Participants
|
|
Race/Ethnicity, Customized
Black
|
11139 Participants
n=31432 Participants
|
11650 Participants
n=33564 Participants
|
22789 Participants
n=64996 Participants
|
|
Race/Ethnicity, Customized
Hispanic
|
7363 Participants
n=31432 Participants
|
8061 Participants
n=33564 Participants
|
15424 Participants
n=64996 Participants
|
|
Race/Ethnicity, Customized
Asian
|
1047 Participants
n=31432 Participants
|
1064 Participants
n=33564 Participants
|
2111 Participants
n=64996 Participants
|
|
Race/Ethnicity, Customized
Other
|
1058 Participants
n=31432 Participants
|
1318 Participants
n=33564 Participants
|
2376 Participants
n=64996 Participants
|
|
Race/Ethnicity, Customized
Unknown
|
301 Participants
n=31432 Participants
|
374 Participants
n=33564 Participants
|
675 Participants
n=64996 Participants
|
|
Region of Enrollment
United States
|
31432 participants
n=31432 Participants
|
33564 participants
n=33564 Participants
|
64996 participants
n=64996 Participants
|
|
Patient Class
Inpatient
|
24231 Participants
n=31432 Participants
|
25992 Participants
n=33564 Participants
|
50223 Participants
n=64996 Participants
|
|
Patient Class
Observation
|
7201 Participants
n=31432 Participants
|
7572 Participants
n=33564 Participants
|
14773 Participants
n=64996 Participants
|
|
Admission Type
Elective
|
13753 Participants
n=31432 Participants
|
14652 Participants
n=33564 Participants
|
28405 Participants
n=64996 Participants
|
|
Admission Type
Emergency
|
17679 Participants
n=31432 Participants
|
18912 Participants
n=33564 Participants
|
36591 Participants
n=64996 Participants
|
|
Elixhauser Comorbidity
|
2.8 comorbidities
STANDARD_DEVIATION 4.8 • n=30668 Participants • Elixhauser Comorbidity was unknown for some participants.
|
2.9 comorbidities
STANDARD_DEVIATION 4.9 • n=32672 Participants • Elixhauser Comorbidity was unknown for some participants.
|
2.9 comorbidities
STANDARD_DEVIATION 4.8 • n=63340 Participants • Elixhauser Comorbidity was unknown for some participants.
|
|
Length of Stay (LOS)
|
5.1 days
STANDARD_DEVIATION 6.5 • n=31432 Participants
|
4.9 days
STANDARD_DEVIATION 6.4 • n=33564 Participants
|
5.0 days
STANDARD_DEVIATION 6.5 • n=64996 Participants
|
|
Insurance
Medicaid
|
8728 Participants
n=31432 Participants
|
8714 Participants
n=33564 Participants
|
17442 Participants
n=64996 Participants
|
|
Insurance
Medicare
|
13111 Participants
n=31432 Participants
|
14364 Participants
n=33564 Participants
|
27475 Participants
n=64996 Participants
|
|
Insurance
Private
|
7438 Participants
n=31432 Participants
|
8038 Participants
n=33564 Participants
|
15476 Participants
n=64996 Participants
|
|
Insurance
Self Pay
|
431 Participants
n=31432 Participants
|
666 Participants
n=33564 Participants
|
1097 Participants
n=64996 Participants
|
|
Insurance
Other
|
204 Participants
n=31432 Participants
|
186 Participants
n=33564 Participants
|
390 Participants
n=64996 Participants
|
|
Insurance
Unknown
|
1520 Participants
n=31432 Participants
|
1596 Participants
n=33564 Participants
|
3116 Participants
n=64996 Participants
|
|
Discharge
Home
|
21176 Participants
n=31432 Participants
|
23335 Participants
n=33564 Participants
|
44511 Participants
n=64996 Participants
|
|
Discharge
Home / Home Health
|
5739 Participants
n=31432 Participants
|
5528 Participants
n=33564 Participants
|
11267 Participants
n=64996 Participants
|
|
Discharge
Skilled Nursing Facility / Rehab
|
2760 Participants
n=31432 Participants
|
2812 Participants
n=33564 Participants
|
5572 Participants
n=64996 Participants
|
|
Discharge
Long Term Acute Care
|
165 Participants
n=31432 Participants
|
171 Participants
n=33564 Participants
|
336 Participants
n=64996 Participants
|
|
Discharge
Other Transfer
|
98 Participants
n=31432 Participants
|
162 Participants
n=33564 Participants
|
260 Participants
n=64996 Participants
|
|
Discharge
AMA
|
412 Participants
n=31432 Participants
|
385 Participants
n=33564 Participants
|
797 Participants
n=64996 Participants
|
|
Discharge
Psych
|
104 Participants
n=31432 Participants
|
93 Participants
n=33564 Participants
|
197 Participants
n=64996 Participants
|
|
Discharge
Other / Unknown
|
50 Participants
n=31432 Participants
|
74 Participants
n=33564 Participants
|
124 Participants
n=64996 Participants
|
|
Discharge
Hospice / Expired
|
928 Participants
n=31432 Participants
|
1004 Participants
n=33564 Participants
|
1932 Participants
n=64996 Participants
|
PRIMARY outcome
Timeframe: 24 monthsThe primary outcome is the proportion of patients who received SBIRT after a positive universal screen for being at risk for substance misuse. The design is an interrupted time-series prospective observational study.
Outcome measures
| Measure |
Usual Care
n=31432 Participants
Before intervention
|
SMART-AI
n=33564 Participants
Automated processing of clinical notes collected during routine care in first 24 hours of hospital admission to identify individuals at-risk for substance misuse to receive standard-of-care full screening and assessment, brief intervention, or referral to treatment (SBIRT) intervention.
Processing of clinical notes in the EHR data collected during routine care: Clinical notes collected in the first day of hospital admission during usual care as input to natural language processing and machine learning algorithm.
|
|---|---|---|
|
Proportion of Patients That Had a Universal Screen Positive and Received SBIRT (Screening, Brief Intervention, or Referral to Treatment)
|
1189 Participants
|
1144 Participants
|
SECONDARY outcome
Timeframe: 12 months enrollment with 6 months follow-up for rehospitalizationWe will compare healthcare utilization outcomes in all patients between pre- and post-periods controlling for all patient demographic and clinical characteristics.
Outcome measures
| Measure |
Usual Care
n=31432 Participants
Before intervention
|
SMART-AI
n=33564 Participants
Automated processing of clinical notes collected during routine care in first 24 hours of hospital admission to identify individuals at-risk for substance misuse to receive standard-of-care full screening and assessment, brief intervention, or referral to treatment (SBIRT) intervention.
Processing of clinical notes in the EHR data collected during routine care: Clinical notes collected in the first day of hospital admission during usual care as input to natural language processing and machine learning algorithm.
|
|---|---|---|
|
All-cause Re-hospitalizations Following 6-months From the Index Hospital Encounter
|
9584 Participants
|
10241 Participants
|
Adverse Events
Usual Care
Serious events: 0 serious events
Other events: 0 other events
Deaths: 0 deaths
SMART-AI: NLP (Natural Language Processing) Pre-screen
Serious events: 0 serious events
Other events: 0 other events
Deaths: 0 deaths
Serious adverse events
Adverse event data not reported
Other adverse events
Adverse event data not reported
Additional Information
Majid Afshar, MD
UW School of Medicine and Public Health
Phone: 608-263-1792
Email: [email protected]
Results disclosure agreements
- Principal investigator is a sponsor employee
- Publication restrictions are in place