Trial Outcomes & Findings for Improving Colorectal Cancer Screening in Racially Diverse Zip Codes Using Navigation and Machine Learning (PCSNaP) (NCT NCT05383976)
NCT ID: NCT05383976
Last Updated: 2026-02-11
Results Overview
Number of patients that participate in the navigation program
Recruitment status
COMPLETED
Target enrollment
385 participants
Primary outcome timeframe
During the three month enrollment period
Results posted on
2026-02-11
Participant Flow
Participant milestones
| Measure |
Patients Residing in 18 Zip Codes in Western and Southwestern Philadelphia
The cohort will consist of patients residing in 18 zip codes in Western and Southwestern Philadelphia who have primary care providers in 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices.
Machine Learning Algorithm with Existing Penn Medicine CRC Patient Navigation Program: This intervention will utilize the existing Penn Medicine Colorectal Cancer patient navigation program. There will be a monthly list of patients with unfilled coloscopies provided, that are risk-stratified according to the machine learning algorithm and select high-risk criteria. The navigation team will prioritize timely outreach and navigation to high-risk patients according to a script that communicates risk.
|
Controls
The cohort will consist of matched controls for the patients in the intervention cohort, also residing in the same 18 zip codes in Western and Southwestern Philadelphia with primary care providers in same 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices. Using clinical and sociodemographic characteristics, we will create matched cohorts of patients between the intervention participants and historical controls from the same clinics in the year prior.
|
|---|---|---|
|
Overall Study
STARTED
|
201
|
184
|
|
Overall Study
COMPLETED
|
199
|
183
|
|
Overall Study
NOT COMPLETED
|
2
|
1
|
Reasons for withdrawal
Withdrawal data not reported
Baseline Characteristics
Improving Colorectal Cancer Screening in Racially Diverse Zip Codes Using Navigation and Machine Learning (PCSNaP)
Baseline characteristics by cohort
| Measure |
Patients Residing in 18 Zip Codes in Western and Southwestern Philadelphia
n=201 Participants
The cohort will consist of patients residing in 18 zip codes in Western and Southwestern Philadelphia who have primary care providers in 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices.
Machine Learning Algorithm with Existing Penn Medicine CRC Patient Navigation Program: This intervention will utilize the existing Penn Medicine CRC patient navigation program. There will be a monthly list of patients with unfilled coloscopies provided, that are risk-stratified according to the machine learning algorithm and select high-risk criteria. The navigation team will prioritize timely outreach and navigation to high-risk patients according to a script that communicates risk.
|
Controls
n=184 Participants
The cohort will consist of matched controls for the patients in the intervention cohort, also residing in the same 18 zip codes in Western and Southwestern Philadelphia with primary care providers in same 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices. Using clinical and sociodemographic characteristics, we will create matched cohorts of patients between the intervention participants and historical controls from the same clinics in the year prior.
|
Total
n=385 Participants
Total of all reporting groups
|
|---|---|---|---|
|
Age, Categorical
<=18 years
|
0 Participants
n=41 Participants
|
0 Participants
n=1581 Participants
|
0 Participants
n=4626 Participants
|
|
Age, Categorical
Between 18 and 65 years
|
135 Participants
n=41 Participants
|
145 Participants
n=1581 Participants
|
280 Participants
n=4626 Participants
|
|
Age, Categorical
>=65 years
|
66 Participants
n=41 Participants
|
39 Participants
n=1581 Participants
|
105 Participants
n=4626 Participants
|
|
Age, Continuous
|
61.28 years
n=41 Participants
|
56.91 years
n=1581 Participants
|
59.10 years
n=4626 Participants
|
|
Sex: Female, Male
Female
|
127 Participants
n=41 Participants
|
116 Participants
n=1581 Participants
|
243 Participants
n=4626 Participants
|
|
Sex: Female, Male
Male
|
74 Participants
n=41 Participants
|
68 Participants
n=1581 Participants
|
142 Participants
n=4626 Participants
|
|
Ethnicity (NIH/OMB)
Hispanic or Latino
|
18 Participants
n=41 Participants
|
16 Participants
n=1581 Participants
|
34 Participants
n=4626 Participants
|
|
Ethnicity (NIH/OMB)
Not Hispanic or Latino
|
183 Participants
n=41 Participants
|
168 Participants
n=1581 Participants
|
351 Participants
n=4626 Participants
|
|
Ethnicity (NIH/OMB)
Unknown or Not Reported
|
0 Participants
n=41 Participants
|
0 Participants
n=1581 Participants
|
0 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
American Indian or Alaska Native
|
0 Participants
n=41 Participants
|
1 Participants
n=1581 Participants
|
1 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
Asian
|
3 Participants
n=41 Participants
|
4 Participants
n=1581 Participants
|
7 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
Native Hawaiian or Other Pacific Islander
|
0 Participants
n=41 Participants
|
0 Participants
n=1581 Participants
|
0 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
Black or African American
|
134 Participants
n=41 Participants
|
134 Participants
n=1581 Participants
|
268 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
White
|
34 Participants
n=41 Participants
|
29 Participants
n=1581 Participants
|
63 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
More than one race
|
1 Participants
n=41 Participants
|
3 Participants
n=1581 Participants
|
4 Participants
n=4626 Participants
|
|
Race (NIH/OMB)
Unknown or Not Reported
|
29 Participants
n=41 Participants
|
13 Participants
n=1581 Participants
|
42 Participants
n=4626 Participants
|
|
Region of Enrollment
United States
|
201 participants
n=41 Participants
|
184 participants
n=1581 Participants
|
385 participants
n=4626 Participants
|
PRIMARY outcome
Timeframe: During the three month enrollment periodPopulation: Population consists of intervention patients who were correctly identified by the machine learning algorithm as needing to complete colorectal cancer screening and were eligible for navigation.
Number of patients that participate in the navigation program
Outcome measures
| Measure |
Patients Residing in 18 Zip Codes in Western and Southwestern Philadelphia
n=119 Participants
The cohort will consist of patients residing in 18 zip codes in Western and Southwestern Philadelphia who have primary care providers in 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices.
Machine Learning Algorithm with Existing Penn Medicine CRC Patient Navigation Program: This intervention will utilize the existing Penn Medicine CRC patient navigation program. There will be a monthly list of patients with unfilled coloscopies provided, that are risk-stratified according to the machine learning algorithm and select high-risk criteria. The navigation team will prioritize timely outreach and navigation to high-risk patients according to a script that communicates risk.
|
Controls
The cohort will consist of matched controls for the patients in the intervention cohort, also residing in the same 18 zip codes in Western and Southwestern Philadelphia with primary care providers in same 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices. Using clinical and sociodemographic characteristics, we will create matched cohorts of patients between the intervention participants and historical controls from the same clinics in the year prior.
|
|---|---|---|
|
Enrollment in Navigator Program (Feasibility)
|
79 Participants
|
—
|
PRIMARY outcome
Timeframe: Within the three month enrollment period and three month follow-up periodPopulation: Population consists of intervention and control patients identified by the machine learning algorithm as needing to complete colorectal cancer screening.
Number of patients that have completed their colonoscopy or Fecal Immunochemical Test (FIT)
Outcome measures
| Measure |
Patients Residing in 18 Zip Codes in Western and Southwestern Philadelphia
n=199 Participants
The cohort will consist of patients residing in 18 zip codes in Western and Southwestern Philadelphia who have primary care providers in 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices.
Machine Learning Algorithm with Existing Penn Medicine CRC Patient Navigation Program: This intervention will utilize the existing Penn Medicine CRC patient navigation program. There will be a monthly list of patients with unfilled coloscopies provided, that are risk-stratified according to the machine learning algorithm and select high-risk criteria. The navigation team will prioritize timely outreach and navigation to high-risk patients according to a script that communicates risk.
|
Controls
n=183 Participants
The cohort will consist of matched controls for the patients in the intervention cohort, also residing in the same 18 zip codes in Western and Southwestern Philadelphia with primary care providers in same 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices. Using clinical and sociodemographic characteristics, we will create matched cohorts of patients between the intervention participants and historical controls from the same clinics in the year prior.
|
|---|---|---|
|
Completion of Colorectal Cancer Screening
|
92 Participants
|
79 Participants
|
PRIMARY outcome
Timeframe: Within the three month enrollment period and three month follow-up periodPopulation: Population consists of patients who completed a colonoscopy during the study period.
Rate of adenomas after completion of colonoscopy
Outcome measures
| Measure |
Patients Residing in 18 Zip Codes in Western and Southwestern Philadelphia
n=59 Participants
The cohort will consist of patients residing in 18 zip codes in Western and Southwestern Philadelphia who have primary care providers in 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices.
Machine Learning Algorithm with Existing Penn Medicine CRC Patient Navigation Program: This intervention will utilize the existing Penn Medicine CRC patient navigation program. There will be a monthly list of patients with unfilled coloscopies provided, that are risk-stratified according to the machine learning algorithm and select high-risk criteria. The navigation team will prioritize timely outreach and navigation to high-risk patients according to a script that communicates risk.
|
Controls
n=52 Participants
The cohort will consist of matched controls for the patients in the intervention cohort, also residing in the same 18 zip codes in Western and Southwestern Philadelphia with primary care providers in same 4 Penn Medicine Internal Medicine practices and 3 Penn Medicine Family Medicine Practices. Using clinical and sociodemographic characteristics, we will create matched cohorts of patients between the intervention participants and historical controls from the same clinics in the year prior.
|
|---|---|---|
|
Number of Participants With Adenoma Detection
|
21 Participants
|
15 Participants
|
Adverse Events
Patients Residing in 18 Zip Codes in Western and Southwestern Philadelphia
Serious events: 0 serious events
Other events: 0 other events
Deaths: 0 deaths
Controls
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
Yvette Frimpong
Abramson Cancer Center at Penn Medicine
Phone: (215) 573-5107
Email: [email protected]
Results disclosure agreements
- Principal investigator is a sponsor employee
- Publication restrictions are in place