Trial Outcomes & Findings for Breaking Implicit Bias Habits: An Individuation Pilot Study In Rheumatology (NCT NCT05116163)

Last Updated: 2025-11-12

Results Overview

Documented receipt of high-quality care at the appointment or within 30 days of the appointment. The quality metrics for Lupus and RA include but are not limited to: Hydroxychloroquine (HCQ) initiation and Folic acid if receiving methotrexate, respectively. These items will be added together to form a score and the number of items completed out of the total will be the primary outcome measure compared across groups. The denominator is the patient-specific quality metrics they are eligible for and the numerator is the number achieved. The final percent ranges from 0-100 (with 100 indicating that 100% of the quality metrics that the patient was eligible for were met).

Recruitment status

COMPLETED

Study phase

Target enrollment

201 participants

Primary outcome timeframe

30 days

Results posted on

2025-11-12

Participant Flow

Rheumatologists were recruited from 6/17/2022-12/12/2023 at Brigham and Women's Hospital and Massachusetts General Hospital.

Rheumatologists took Implicit Association Tests and were provided with implicit bias training and then randomized to either the control or intervention arm.

Participant milestones

Participant milestones
Measure	Arm 1: Rheumatologist Individuation Intervention Plus Implicit Bias Education Rheumatologists were randomized to this arm. They first watched implicit bias educational modules and then were instructed to incorporate an individuation strategy into each of their clinical encounters. Rheumatologists were given a choice of several individuation-related questions to better understand the unique characteristics of each patient. Once the provider decided on his/her choice phrases, the research team assisted with the development of a smart phrase (also called "dot phrase") to allow them to incorporate this into a note. Once a week, rheumatologists received an email reminding them to incorporate this question and the documentation into their encounters.	Arm 2: Rheumatologist Implicit Bias Education Only Rheumatologists were randomized to this arm. They watched implicit bias educational modules (same as those used in Arm 1) and received no further intervention.	Patients of Arm 1 Providers- Chart Review Only Patients of providers randomized to Arm 1 (individuation intervention). These patients were not intervened upon; their charts from routine care were reviewed only if they met inclusion criteria.	Patients of Arm 2 Providers- Chart Review Only Patients of providers randomized to Arm 2. These patients were not intervened upon; their charts from routine care were reviewed only if they met inclusion criteria.	Patients of Arm 1 Rheumatologists Enrolled for Audiorecording Patients of rheumatologists randomized to Arm 1 who consented for audiorecording and completed surveys, and participated in chart reviews.	Patients of Arm 2 Rheumatologists Enrolled for Audiorecording Patients of rheumatologists randomized to Arm 2 who consented for audiorecording and completed surveys, and participated in chart reviews.
Overall Study STARTED	10	9	81	54	20	27
Overall Study COMPLETED	10	9	79	50	19	27
Overall Study NOT COMPLETED	0	0	2	4	1	0

Reasons for withdrawal

Withdrawal data not reported

Baseline Characteristics

This analysis was only completed for providers; therefore the number analyzed for each of the patient categories is 0.

Baseline characteristics by cohort

Baseline characteristics by cohort
Measure	Arm 1: Individuation Intervention Plus Implicit Bias Education: Rheumatologists n=10 Participants Rheumatologists watched implicit bias training modules and then were instructed to incorporate an individuation strategy into each of their clinical encounters.	Arm 2: Implicit Bias Education Only: Rheumatologists n=9 Participants Rheumatologists watched implicit bias training modules.	Patients of Arm 1- Chart Review Only n=79 Participants Patients of providers in Arm 1 who had their electronic health records reviewed but did not enroll/complete surveys	Patients of Arm 2- Chart Review Only n=50 Participants Patients of providers in Arm 2 who had their electronic health records reviewed but did not enroll/complete surveys	Patients of Arm 1 Providers With Audiorecordings and Surveys n=20 Participants Patients of providers in Arm 1 who enrolled for an audiorecorded encounter and surveys	Patients of Arm 2 Providers With Audiorecordings and Surveys n=27 Participants Patients of providers in Arm 2 who enrolled for an audiorecorded encounter and surveys	Total n=195 Participants Total of all reporting groups
Sex/Gender, Customized Female	5 Participants n=10 Participants	5 Participants n=9 Participants	53 Participants n=79 Participants	37 Participants n=50 Participants	11 Participants n=20 Participants	17 Participants n=27 Participants	128 Participants n=195 Participants
Sex/Gender, Customized Male	5 Participants n=10 Participants	4 Participants n=9 Participants	6 Participants n=79 Participants	4 Participants n=50 Participants	4 Participants n=20 Participants	2 Participants n=27 Participants	25 Participants n=195 Participants
Age, Categorical <=18 years	0 Participants n=10 Participants	0 Participants n=9 Participants	0 Participants n=79 Participants	0 Participants n=50 Participants	0 Participants n=20 Participants	0 Participants n=27 Participants	0 Participants n=195 Participants
Age, Categorical Between 18 and 65 years	9 Participants n=10 Participants	8 Participants n=9 Participants	53 Participants n=79 Participants	38 Participants n=50 Participants	17 Participants n=20 Participants	23 Participants n=27 Participants	148 Participants n=195 Participants
Age, Categorical >=65 years	1 Participants n=10 Participants	1 Participants n=9 Participants	26 Participants n=79 Participants	12 Participants n=50 Participants	3 Participants n=20 Participants	4 Participants n=27 Participants	47 Participants n=195 Participants
Sex/Gender, Customized Unknown	0 Participants n=10 Participants	0 Participants n=9 Participants	20 Participants n=79 Participants	9 Participants n=50 Participants	5 Participants n=20 Participants	8 Participants n=27 Participants	42 Participants n=195 Participants
Ethnicity (NIH/OMB) Hispanic or Latino	0 Participants n=10 Participants	3 Participants n=9 Participants	9 Participants n=79 Participants	10 Participants n=50 Participants	2 Participants n=20 Participants	5 Participants n=27 Participants	29 Participants n=195 Participants
Ethnicity (NIH/OMB) Not Hispanic or Latino	9 Participants n=10 Participants	6 Participants n=9 Participants	67 Participants n=79 Participants	38 Participants n=50 Participants	17 Participants n=20 Participants	21 Participants n=27 Participants	158 Participants n=195 Participants
Ethnicity (NIH/OMB) Unknown or Not Reported	1 Participants n=10 Participants	0 Participants n=9 Participants	3 Participants n=79 Participants	2 Participants n=50 Participants	1 Participants n=20 Participants	1 Participants n=27 Participants	8 Participants n=195 Participants
Race (NIH/OMB) American Indian or Alaska Native	0 Participants n=10 Participants	0 Participants n=9 Participants	0 Participants n=79 Participants	1 Participants n=50 Participants	0 Participants n=20 Participants	0 Participants n=27 Participants	1 Participants n=195 Participants
Race (NIH/OMB) Asian	1 Participants n=10 Participants	2 Participants n=9 Participants	0 Participants n=79 Participants	0 Participants n=50 Participants	1 Participants n=20 Participants	0 Participants n=27 Participants	4 Participants n=195 Participants
Race (NIH/OMB) Native Hawaiian or Other Pacific Islander	0 Participants n=10 Participants	0 Participants n=9 Participants	0 Participants n=79 Participants	0 Participants n=50 Participants	0 Participants n=20 Participants	0 Participants n=27 Participants	0 Participants n=195 Participants
Race (NIH/OMB) Black or African American	0 Participants n=10 Participants	1 Participants n=9 Participants	61 Participants n=79 Participants	37 Participants n=50 Participants	14 Participants n=20 Participants	23 Participants n=27 Participants	136 Participants n=195 Participants
Race (NIH/OMB) White	8 Participants n=10 Participants	6 Participants n=9 Participants	10 Participants n=79 Participants	6 Participants n=50 Participants	3 Participants n=20 Participants	4 Participants n=27 Participants	37 Participants n=195 Participants
Race (NIH/OMB) More than one race	0 Participants n=10 Participants	0 Participants n=9 Participants	2 Participants n=79 Participants	1 Participants n=50 Participants	0 Participants n=20 Participants	0 Participants n=27 Participants	3 Participants n=195 Participants
Race (NIH/OMB) Unknown or Not Reported	1 Participants n=10 Participants	0 Participants n=9 Participants	6 Participants n=79 Participants	5 Participants n=50 Participants	2 Participants n=20 Participants	0 Participants n=27 Participants	14 Participants n=195 Participants
Region of Enrollment United States	10 participants n=10 Participants	9 participants n=9 Participants	79 participants n=79 Participants	50 participants n=50 Participants	20 participants n=20 Participants	27 participants n=27 Participants	248 participants n=195 Participants
Provider-Only Implicit Association Test- Race Bias	-0.46 units on a scale STANDARD_DEVIATION 0.54 • n=10 Participants • This analysis was only completed for providers; therefore the number analyzed for each of the patient categories is 0.	-0.37 units on a scale STANDARD_DEVIATION 0.57 • n=9 Participants • This analysis was only completed for providers; therefore the number analyzed for each of the patient categories is 0.	—	—	—	—	-0.42 units on a scale STANDARD_DEVIATION 0.54 • n=19 Participants • This analysis was only completed for providers; therefore the number analyzed for each of the patient categories is 0.
Provider-Only Implicit Association Tests - Race stereotyping	-0.13 units on a scale STANDARD_DEVIATION 0.76 • n=10 Participants • Provider only surveys	-0.29 units on a scale STANDARD_DEVIATION 0.37 • n=9 Participants • Provider only surveys	—	—	—	—	-0.20 units on a scale STANDARD_DEVIATION 0.60 • n=19 Participants • Provider only surveys
Provider-Only Implicit Association Test - Socioeconomic status bias	0.75 units on a scale STANDARD_DEVIATION 0.55 • n=10 Participants • Provider only surveys	0.54 units on a scale STANDARD_DEVIATION 0.35 • n=9 Participants • Provider only surveys	—	—	—	—	0.65 units on a scale STANDARD_DEVIATION 0.47 • n=19 Participants • Provider only surveys
Provider-Only Implicit Association Test - Socioeconomic status stereotyping	0.54 units on a scale STANDARD_DEVIATION 0.60 • n=10 Participants • Provider only surveys	0.48 units on a scale STANDARD_DEVIATION 0.49 • n=9 Participants • Provider only surveys	—	—	—	—	0.51 units on a scale STANDARD_DEVIATION 0.54 • n=19 Participants • Provider only surveys
Baseline Medication Adherence	—	—	—	—	49.3 units on a scale STANDARD_DEVIATION 19.6 • n=20 Participants • Enrolled patient-only measure	58.9 units on a scale STANDARD_DEVIATION 18.8 • n=27 Participants • Enrolled patient-only measure	53.8 units on a scale STANDARD_DEVIATION 19.5 • n=47 Participants • Enrolled patient-only measure

PRIMARY outcome

Timeframe: 30 days

Population: Patients of providers randomized to Arm 1 or Arm 2, which includes the eligible patients in the chart-review groups, and those who also participated in the audiorecordings and surveys

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=98 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=77 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Percent of Quality Metrics Achieved	62.14 Percent of Quality Metrics Achieved Standard Deviation 13.46	54.31 Percent of Quality Metrics Achieved Standard Deviation 14.46

SECONDARY outcome

Timeframe: At time of appointment (baseline) during which audiorecording took place, reflecting the experience at the appointment immediately preceding survey completion.

Population: Subset of patients who consented to participate in audiorecordings and returned completed surveys.

Differences in perception of patient centeredness comparing the intervention to non-intervention group. Higher scores indicate less patient-centered experiences. 4-point Likert scale, score range from 14-56.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=19 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=21 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Perception of Patient Centeredness	27.5 score on a scale Standard Deviation 9	23.4 score on a scale Standard Deviation 7

SECONDARY outcome

Timeframe: At time of appointment (baseline) during which audiorecording took place, reflecting the experience at the appointment immediately preceding survey completion.

Population: Subset of patients who consented to participate in audiorecordings and returned completed surveys.

Differences in patient satisfaction score, comparing the intervention to non-intervention group. Score range from 20-70. Higher scores indicate greater patient satisfaction.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=17 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=22 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Patient Satisfaction	50.4 score on a scale Standard Deviation 9	51.5 score on a scale Standard Deviation 8

SECONDARY outcome

Timeframe: A time of the appointment when audiorecording took place (baseline) asking patients to describe the degree to which each item occurs in their day-to-day life.

Population: Subset of patients who consented to participate in audiorecordings and returned completed surveys.

Reflects experiences of everyday discrimination, comparing the intervention to non-intervention group. Score range is 0-40. Higher scores indicate a greater amount of discrimination encountered.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=19 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=25 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Everyday Discrimination Scale	0.96 score on a scale Standard Deviation 0.84	1.16 score on a scale Standard Deviation 0.93

SECONDARY outcome

Timeframe: 3 months after the start of the intervention

Population: Subset of patients in the audiorecording/survey arms who returned completed 3 month follow up adherence surveys.

Adherence score (for rheumatology medications), comparing the 3-month adherence score to baseline for the intervention to non-intervention group at three months following the encounter date. Score range is 0-100 with zero being the lowest adherence, and 100 being the best.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=13 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=12 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Adherence	54.96 score on a scale Standard Deviation 14.15	55.99 score on a scale Standard Deviation 19.44

SECONDARY outcome

Timeframe: One time baseline evaluation from the transcript of the audiorecording of the encounter of the patient with their rheumatologist

Population: Patients who participated in audiorecordings and had transcripts for analysis.

Provider communication will be measured using the recorded transcripts and compared between arms. We compared the mean (SD) of provider positive emotion words expressed during a clinical encounter. Measured only for patients enrolled in the study who had their appointment audiorecorded.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=19 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=27 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Provider Communication: Positive Emotion Words	0.99 Number of words Standard Deviation 0.50	0.66 Number of words Standard Deviation 0.33

SECONDARY outcome

Timeframe: One time baseline evaluation after the appointment reflecting experiences at the immediately preceding appointment.

Population: Subset of patients who consented to participate in audiorecordings and returned completed surveys.

Patient trust in their providers, comparing the intervention to non-intervention group. Responses are summed (range 5-25) with higher scores indicating more trust.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=19 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=24 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Patient Trust	16.4 score on a scale Standard Deviation 4	17.6 score on a scale Standard Deviation 5

SECONDARY outcome

Timeframe: 6 months following the encounter of interest

Comparison of incidence rate of Emergency Department visits between patients of providers in the intervention compared to the control arms.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=98 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=77 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Incidence Rate Ratio of Emergency Departments Visits	0.398 number of ED visits Standard Deviation 0.971	0.390 number of ED visits Standard Deviation 0.934

SECONDARY outcome

Timeframe: 6 months following the encounter of interest

Comparison of incidence rate of outpatient visits between patients of providers in the intervention compared to the control arms.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=98 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=77 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Incidence Rate Ratio of Outpatient Visits.	7.34 Number of outpatient visits Standard Deviation 6.22	6.05 Number of outpatient visits Standard Deviation 5.88

SECONDARY outcome

Timeframe: 6 months after physician encounter

Comparison of incidence rate of hospitalizations between patients of providers in the intervention compared to the control arms.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=98 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=77 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Incidence Rate Ratio of Hospitalizations	0.316 number of hospitalizations Standard Deviation 1.06	0.779 number of hospitalizations Standard Deviation 0.354

SECONDARY outcome

Timeframe: Baseline compared to 6 months after the start of the intervention.

Change in provider IAT scores pre and post intervention. IAT scores range from -1 to 1. A score of 0 indicates no strong implicit bias. A score of 1 is pro-high socioeconomic status bias and a score of -1 is pro-low socioeconomic status bias.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=10 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=9 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Implicit Association Test (IAT) Scores- Socioeconomic Status Stereotyping	0.50 units on a scale Standard Deviation 0.37	0.15 units on a scale Standard Deviation 0.50

SECONDARY outcome

Timeframe: Baseline compared to 6 months after the start of the intervention.

Change in provider IAT scores pre and post intervention. IAT scores range from -1 to 1. A score of 0 indicates no strong implicit bias. A score of 1 is pro-White bias and a score of -1 is pro-Black bias.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=10 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=9 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Implicit Association Test (IAT) Scores- Race Implicit Bias	-0.27 units on a scale Standard Deviation 0.47	0.12 units on a scale Standard Deviation 0.44

SECONDARY outcome

Timeframe: Baseline compared to 6 months after the start of the intervention.

Change in provider IAT scores pre and post intervention. IAT scores range from -1 to 1. A score of 0 indicates no strong implicit bias. A score of 1 is pro-White stereotyping and a score of -1 is pro-Black stereotyping.

Outcome measures

Outcome measures
Measure	Patients of Arm 1 Providers Enrolled for Audiorecording and Surveys n=10 Participants Subset of patients of providers in Arm 1 who enrolled for audiorecordings and completed surveys	Patients of Arm 2 Providers Enrolled for Audiorecording and Surveys n=9 Participants Subset of patients of providers in Arm 2 who enrolled in audiorecording and completed surveys
Implicit Association Test (IAT) Scores- Race Stereotyping.	-0.25 units on a scale Standard Deviation 0.61	-0.39 units on a scale Standard Deviation 0.50

SECONDARY outcome

Timeframe: Baseline compared to 6 months after the start of the intervention.

. Change in provider IAT scores pre and post intervention. IAT scores range from -1 to 1. A score of 0 indicates no strong implicit bias. A score of 1 is pro-high socioeconomic status bias and a score of -1 is pro-low socioeconomic status bias.