P2370 - Determining Patient Preferences for Colorectal Cancer Screening Tests Using Conjoint Analysis

Tuesday, October 26, 2021

10:30 AM – 4:00 PM PT

Location: Shoreline Exhibit Hall

Has Audio

Presenting Author(s)

Katherine Makaroff, BA

Cedars-Sinai Center for Outcomes Research and Education
Los Angeles, CA, United States

Katherine Makaroff, BA¹, Jaspreet Shergill, MD, MS², Marie Lauzon, MS², Brennan Spiegel, MD, MSHS², Christopher V. Almario, MD, MSHPM, FACG²
¹Cedars-Sinai Center for Outcomes Research and Education, Los Angeles, CA; ²Cedars-Sinai Medical Center, Los Angeles, CA

Introduction: As the US Preventive Services Task Force recently recommended to start screening for CRC at age 45, it is critical to update our understanding on people’s preferences for the different screening tests. In this study, we examined individual decision making surrounding the US Multi-Society Task Force (MSTF)-recommended tests using conjoint analysis—a form of tradeoff analysis that determines how people make complex decisions.

Methods: To quantify people’s preferences for CRC screening tests, we conducted a choice-based conjoint analysis survey in individuals ≥40yo at average risk for CRC who had not undergone prior screening. We recruited a nationwide sample via Cint (4/29-6/2/21), using Census quotas for sex and region. The hypothetical screening tests presented during the conjoint were modeled on the five MSTF-recommended options (Figure). Using the conjoint data, we determined the preferred test for each individual when considering: (i) all five MSTF-recommended tests; (ii) only Tier 1-recommended tests (FIT or colonoscopy). We then performed logistic regression to explore whether patient demographics, comorbidities, or beliefs on CRC predicted decision making on FIT vs colonoscopy.

Results: Overall, 1,000 participants completed the survey, and the Figure shows their preferred tests. When considering all five MSTF-recommended tests, most preferred a FIT-DNA, followed by colon video capsule, FIT, colonoscopy, and CT colonography. When considering only the Tier 1 tests, 73.3% preferred FIT while 26.7% preferred a colonoscopy. In regression analysis, demographics and comorbidities largely did not predict whether respondents preferred FIT over colonoscopy (Table). However, respondents ≥60yo alongside those with more perceived barriers to screening were more likely to prefer FIT. Individuals planning to get screened and those who believed they were more susceptible to CRC were more likely to prefer colonoscopy.

Discussion: When considering the MSTF Tier 1-recommended tests, we found that 3 in 4 individuals preferred FIT over colonoscopy. We also noted that demographics and clinical factors are poor predictors of decision making. Thus, to significantly increase CRC screening uptake, organized screening programs should consider a sequential testing approach (offer FIT first, if declines offer colonoscopy). In opportunistic settings, clinicians should describe the screening options either sequentially or as a choice (FIT vs colonoscopy).

Figure: Figure. Data from simulations using conjoint analysis data assessing the proportion of respondents who would prefer each CRC screening test (N=1,000).

Variable	Prefers FIT every year for CRC screening
Variable	n (% of row) or median (IQR)	aOR [95% CI]
Age:
40-49 yo	301 (66.0)	reference
50-59 yo	198 (74.2)	1.38 [0.96-2.00]
≥60 yo	234 (84.5)	2.11 [1.34-3.32]
Sex:
Male	330 (67.8)	reference
Female	402 (78.5)	1.38 [0.99-1.91]
Prefer not to say	1 (100.0)	---
Race/ethnicity:
Non-Hispanic White	593 (73.6)	reference
Non-Hispanic Black	45 (70.3)	0.79 [0.43-1.44]
Latinx	39 (68.4)	0.93 [0.49-1.75]
Non-Hispanic Asian	31 (75.6)	1.19 [0.54-2.63]
Other	25 (78.1)	1.00 [0.41-2.47]
Educational attainment:
High school degree or less	216 (76.9)	reference
Some college education	190 (76.6)	0.97 [0.62-1.51]
College degree	247 (71.2)	0.77 [0.51-1.17]
Graduate degree	80 (64.5)	0.75 [0.43-1.30]
Marital status:
Married or living with a partner	373 (70.5)	reference
Not married	360 (76.4)	1.10 [0.78-1.54]
Total household income, $:
< 50,000	396 (76.6)	reference
50,000–100,000	224 (75.2)	1.17 [0.78-1.73]
≥100,001	93 (57.8)	0.76 [0.46-1.25]
Prefer not to say	20 (83.3)	1.44 [0.44-4.71]
Employment status:
Unemployed, on disability, on leave of absence from work, retired, or a homemaker	364 (77.3)	reference
Employed or student	369 (69.8)	1.18 [0.82-1.68]
Has health insurance	614 (72.7)	1.05 [0.65-1.68]
Has usual source of care	574 (71.8)	0.89 [0.57-1.37]
Self-reported health status:
Excellent	64 (72.7)	reference
Very good	197 (70.4)	0.97 [0.53-1.75]
Good	301 (74.0)	1.25 [0.69-2.27]
Fair	145 (75.5)	1.26 [0.65-2.47]
Poor	26 (78.8)	1.51 [0.51-4.50]
Number of medical comorbidities ^a:
0	193 (76.6)	reference
1	169 (75.1)	1.08 [0.68-1.72]
2	145 (68.7)	0.75 [0.47-1.21]
≥3	226 (72.4)	0.87 [0.54-1.40]
Number of GI comorbidities ^b:
0	556 (73.2)	reference
1	127 (74.7)	1.03 [0.67-1.57]
2	36 (66.7)	0.84 [0.43-1.64]
≥3	14 (87.5)	2.50 [0.51-12.30]
Number of GI symptoms experienced in past 3 months ^c:
0	365 (76.2)	reference
1	139 (68.8)	0.88 [0.59-1.31]
2	98 (73.7)	1.07 [0.65-1.75]
≥3	131 (70.4)	0.89 [0.56-1.41]
US region:
Northeast	169 (75.1)	reference
South	254 (73.8)	1.34 [0.87-2.08]
Midwest	142 (68.6)	0.95 [0.59-1.51]
West	168 (75.0)	1.15 [0.72-1.83]
Has non-first degree relative or friend diagnosed with CRC	96 (69.6)	1.02 [0.66-1.58]
Plans to get screened for CRC	332 (64.0)	0.55 [0.38-0.79]
Self-perceived CRC susceptibility (1-5 scale; higher=more susceptible)	2.4 [1.2]	0.75 [0.61-0.93]
Self-perceived impact of CRC diagnosis (1-5 scale; higher=more severe impact)	3.2 [0.8]	0.89 [0.68-1.15]
Self-perceived benefits of CRC screening (1-5 scale; higher=more beneficial)	4.0 [0.8]	0.90 [0.67-1.19]
Self-perceived barriers to CRC screening (1-5 scale; higher=more barriers)	2.8 [0.7]	1.34 [1.03-1.74]
Note: All the variables in the table were included in the multivariable logistic regression model. Sex–prefer not to say was not included in the regression model as it perfectly predicted the outcome. a Includes migraines, heart disease, high blood pressure, lung disease, diabetes, kidney disease, anemia or other blood disease, cancer, depression, osteoarthritis or degenerative arthritis, back pain, rheumatoid arthritis, or other medical problems. b Includes irritable bowel syndrome, ulcer or stomach disease, liver disease, celiac disease, cirrhosis, diverticulitis, gallstones, gastroenteritis, gastroesophageal reflux disease, gastroparesis, or pancreatitis. c Includes abdominal pain or discomfort, anal or rectal pain, bloating, bowel incontinence, constipation, diarrhea, dysphagia, heartburn, nausea/vomiting, or regurgitation.

Table: Table. Regression analysis on preferring FIT every year over colonoscopy every 10 years for CRC screening; screening test preferences were determined through simulations from conjoint analysis-derived data (Nf1,000).

Disclosures:
Katherine Makaroff indicated no relevant financial relationships.
Jaspreet Shergill indicated no relevant financial relationships.
Marie Lauzon indicated no relevant financial relationships.
Brennan Spiegel indicated no relevant financial relationships.
Christopher Almario indicated no relevant financial relationships.

Katherine Makaroff, BA¹, Jaspreet Shergill, MD, MS², Marie Lauzon, MS², Brennan Spiegel, MD, MSHS², Christopher V. Almario, MD, MSHPM, FACG². P2370 - Determining Patient Preferences for Colorectal Cancer Screening Tests Using Conjoint Analysis, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.