Colorectal Cancer Screening
Conditions
Keywords
decision aid, cancer screening, colorectal neoplasm, decision making, risk communication, numeracy, health literacy
Brief summary
Experts believe that increasing the low uptake of screening for colorectal cancer (CRC) requires educating patients about all approved tests and helping them choose one that fits their preferences. As one motto puts it: The best test is the one that gets done. Screening tests range from more invasive and very sensitive for polyps and cancer (colonoscopy) to less invasive and less sensitive (e.g., fecal immunochemical testing (FIT)). But it is unclear how best to educate patients about the options and the tradeoffs involved. Some guidelines recommend that decision aids, a promising tool in this area, provide patients with detailed quantitative information, including baseline risk, risk reduction, and chance of negative outcomes. But this sort of comparative effectiveness data can confuse patients, especially those with limited mathematical ability. Previous studies have not measured the effect of providing quantitative information to patients with varying levels of ability or interest or asked them whether such data is essential for their decision-making. The investigators will conduct a clinical trial to determine the impact on patients who view a decision aid (DA) that includes quantitative information versus a DA without such data. The investigators will also seek to determine whether numeracy moderates the effect of quantitative information.
Interventions
BEHAVIORALDA - Quantitative
BEHAVIORALDA - Verbal
Sponsors
Patient-Centered Outcomes Research Institute
Indiana University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SCREENING
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
50 Years to 75 Years
Healthy volunteers
No
Inclusion criteria
1. Have not had colonoscopy performed in last 10 years, sigmoidoscopy in last 5 years, or fecal occult blood testing (including FIT) in last 1 year and 2. have a scheduled appointment or due to schedule an appointment with a healthcare practitioner at our performance sites.
Exclusion criteria
1. undergoing workup for symptoms consistent with colon cancer, such as weight loss or rectal bleeding 2. have a diagnosis or medical history conferring elevated risk for CRC including polypectomy or colon cancer, inflammatory bowel disease, certain inherited syndromes, or a significant family history of CRC 3. are unable to speak and read English.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Colorectal Cancer (CRC) Screening Completion | 6 months post intervention | Completion of colonoscopy, fecal immunochemical testing (FIT), or other CRC screening test within 6 months of enrollment, based on documentation in the participants' electronic health record. If a patient completed both screening tests and the colonoscopy was a follow-up to a positive FIT, they were considered having completed a FIT only. If the FIT was negative and the patient still had a colonoscopy, they were considered to have completed both tests. |
| Colorectal Cancer (CRC) Screening Intention: Change From Baseline to Post-intervention | 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid]) | Multiple choice question assessing subject's intention in getting a CRC screening test in the next 6 months. The response options were: 5=Definitely, 4=Probably, 3=May or may not, 2=Probably not, and 1=Definitely not. Higher positive change means CRC screening intention increased. Increased intention to be screened is a better outcome. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Perceived Barriers to Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) or Colonoscopy: Change From Baseline to Post-intervention | 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid]) | Assessed separately for colonoscopy (11 questions) and FIT (9 questions), using items drawn from validated scales for measuring barriers and self-efficacy of colonoscopy and FIT. All items had Likert-type response options where 5=strongly agree to 1=strongly disagree. Mean values were calculated within each set of questions (each combined score has a range of 1 to 5) at baseline and post-intervention and then change from baseline was calculated using the mean at post-intervention minus the mean at baseline. The range for change could be -4 to 4 with lower values meaning perceived barriers decreased which is a better outcome. |
| Knowledge of Colorectal Cancer (CRC) and Colorectal Cancer Screening: Change in Number of Correct Answers From Baseline to Post-intervention | 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid]) | Qualitative knowledge is assessed with five multiple choice and five true/false questions regarding general information (including risk factors, screening test options, and test frequency) of CRC and CRC screening. Knowledge scores were derived by summing correct responses to the 10 individual knowledge questions (range, 0-10), and change from baseline was calculated using the knowledge score at post-intervention minus the knowledge score at baseline. The range for change could be 0 to 10 with higher values meaning an increase in knowledge which is a better outcome. |
| Perceived Risk of Colorectal Cancer (CRC): Change From Baseline to Post-intervention | 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid]) | Multiple choice questions assessing subject's perception of how likely they are to get colon cancer in the next 5 years, in the next 10 years, and sometime during their lifetime. Each had response options: 4=very likely, 3=somewhat likely, 2=somewhat unlikely, and 1=very unlikely. The mean of the 3 questions was calculated (each combined score has a range of 1 to 4) at baseline and post-intervention and then change from baseline was calculated using the mean at post-intervention minus the mean at baseline. The range for change could be from -3 to 3 with higher values meaning an increase in perceived risk which is a better outcome. |
| Number of Patients With High and Low Subjective Numeracy | 1 day | Subjective numeracy was determined using the Subjective Numeracy Scale (SNS), a validated instrument that involves 8 Likert-type questions regarding the participant's preference for or dislike of numeric information and their perception of their ability to understand it. Each item has a six-point response option from 1-6. The higher the number, the higher the subjective numeracy. All participants were divided into two groups: above (high numeracy) and below (low numeracy) the median for their total subjective numeracy score. |
| Decision Conflict: Change in Conflict From Baseline to Post-intervention | 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid]) | Decision conflict is assessed using the16-item Decision Conflict Scale. Scores will range from 0 (low decision conflict) to 100 (high decision conflict). Negative change means decision conflict decreased. A decrease in decision conflict is a better outcome. |
| Perceived Benefits of Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) and Colonoscopy: Change From Baseline to Post-intervention | 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid]) | Assessed separately for colonoscopy (4 questions) and FIT (4 questions), using items drawn from validated scales for measuring benefits and self-efficacy of colonoscopy and FIT. All items had Likert-type response options where 5=strongly agree to 1=strongly disagree. Mean values were calculated within each set of questions (each combined score has a range of 1 to 5) at baseline and post-intervention and then change from baseline was calculated using the mean at post-intervention minus the mean at baseline. The range for change could be -4 to 4 with higher values meaning perceived benefits increased which is a better outcome. |
Countries
United States
Participant flow
Recruitment details
Participants were recruited from primary care practices between June 2015 and December 2016.
Participants by arm
| Arm | Count |
|---|---|
| Verbal Subjects view:
1. A computer-based presentation regarding colorectal cancer (CRC) and screening for CRC with colonoscopy and stool testing. Includes excerpts from a video from the American Cancer Society.
2. Computer-based presentation providing verbal information regarding (a) the lifetime average probability of getting and dying from CRC, (b) the reduction in morbidity and mortality provided by colonoscopy and stool tests, (c) the sensitivity of colonoscopy and stool tests for finding cancer, (d) the false negative values of colonoscopy and stool tests for finding cancer, (e) the risk of heavy bleeding or colon injury from colonoscopy, and (f) the chance of having a positive stool test.
DA - Verbal | 344 |
| Quantitative Subjects view:
1. A computer-based presentation regarding colorectal cancer (CRC) and screening for CRC with colonoscopy and stool testing. Includes excerpts from a video from the American Cancer Society.
2. Computer-based presentation providing quantitative information regarding (a) the lifetime average probability of getting and dying from CRC, (b) the reduction in morbidity and mortality provided by colonoscopy and stool tests, (c) the sensitivity of colonoscopy and stool tests for finding cancer, (d) the false negative values of colonoscopy and stool tests for finding cancer, (e) the risk of heavy bleeding or colon injury from colonoscopy, and (f) the chance of having a positive stool test.
DA - Quantitative | 344 |
| Total | 688 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Elevated risk for CRC | 2 | 2 |
| Overall Study | MD advised against screening | 1 | 3 |
| Overall Study | No MD visit | 1 | 1 |
| Overall Study | patient deceased | 1 | 0 |
| Overall Study | Screening up-to-date | 12 | 13 |
| Overall Study | Technical issues with intervention | 2 | 0 |
| Overall Study | Workup consistent w/CRC | 1 | 1 |
Baseline characteristics
| Characteristic | Verbal | Total | Quantitative |
|---|---|---|---|
| Age, Continuous | 58.71 years STANDARD_DEVIATION 6.49 | 58.77 years STANDARD_DEVIATION 6.57 | 58.83 years STANDARD_DEVIATION 6.66 |
| Education Associate's degree | 35 Participants | 66 Participants | 31 Participants |
| Education Bachelor's degree | 58 Participants | 126 Participants | 68 Participants |
| Education High school graduate/GED | 75 Participants | 164 Participants | 89 Participants |
| Education Less than high school | 25 Participants | 40 Participants | 15 Participants |
| Education Professional or graduate degree | 44 Participants | 95 Participants | 51 Participants |
| Education Some college/technical school/trade school | 92 Participants | 172 Participants | 80 Participants |
| Education Unknown or Not Reported | 15 Participants | 25 Participants | 10 Participants |
| Employed? No | 170 Participants | 316 Participants | 146 Participants |
| Employed? Unknown or Not Reported | 16 Participants | 26 Participants | 10 Participants |
| Employed? Yes | 158 Participants | 346 Participants | 188 Participants |
| Ethnicity (NIH/OMB) Hispanic or Latino | 12 Participants | 17 Participants | 5 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 317 Participants | 646 Participants | 329 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 15 Participants | 25 Participants | 10 Participants |
| Health Insurance? No | 13 Participants | 30 Participants | 17 Participants |
| Health Insurance? Unknown or Not Reported | 16 Participants | 26 Participants | 10 Participants |
| Health Insurance? Yes | 315 Participants | 632 Participants | 317 Participants |
| Income Level Subjective Are comfortable | 129 Participants | 274 Participants | 145 Participants |
| Income Level Subjective Do not have enough to make ends meet | 80 Participants | 139 Participants | 59 Participants |
| Income Level Subjective Have just enough to make ends meet | 117 Participants | 242 Participants | 125 Participants |
| Income Level Subjective Unknown or Not Reported | 18 Participants | 33 Participants | 15 Participants |
| Income Level (yearly) $20,000 - $39,999 | 59 Participants | 132 Participants | 73 Participants |
| Income Level (yearly) $40,000 - $59,000 | 37 Participants | 77 Participants | 40 Participants |
| Income Level (yearly) $60,000 - $99,999 | 67 Participants | 130 Participants | 63 Participants |
| Income Level (yearly) Less than $20,000 | 95 Participants | 175 Participants | 80 Participants |
| Income Level (yearly) More than $100,000 | 47 Participants | 103 Participants | 56 Participants |
| Income Level (yearly) Unknown or Not Reported | 39 Participants | 71 Participants | 32 Participants |
| Insurance Lapse over past 12 months? No | 278 Participants | 564 Participants | 286 Participants |
| Insurance Lapse over past 12 months? Unknown or Not Reported | 15 Participants | 27 Participants | 12 Participants |
| Insurance Lapse over past 12 months? Yes | 51 Participants | 97 Participants | 46 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 1 Participants | 2 Participants | 1 Participants |
| Race (NIH/OMB) Asian | 3 Participants | 7 Participants | 4 Participants |
| Race (NIH/OMB) Black or African American | 98 Participants | 184 Participants | 86 Participants |
| Race (NIH/OMB) More than one race | 16 Participants | 21 Participants | 5 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 1 Participants | 2 Participants | 1 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 20 Participants | 35 Participants | 15 Participants |
| Race (NIH/OMB) White | 205 Participants | 437 Participants | 232 Participants |
| Relationship Divorced | 61 Participants | 134 Participants | 73 Participants |
| Relationship Married/living with partner | 170 Participants | 338 Participants | 168 Participants |
| Relationship Seperated | 12 Participants | 22 Participants | 10 Participants |
| Relationship Single | 71 Participants | 146 Participants | 75 Participants |
| Relationship Unknown or Not Reported | 11 Participants | 21 Participants | 10 Participants |
| Relationship Widowed | 19 Participants | 27 Participants | 8 Participants |
| Self-reported health status at baseline Excellent | 23 Participants | 56 Participants | 33 Participants |
| Self-reported health status at baseline Fair | 63 Participants | 114 Participants | 51 Participants |
| Self-reported health status at baseline Good | 144 Participants | 277 Participants | 133 Participants |
| Self-reported health status at baseline Poor | 11 Participants | 17 Participants | 6 Participants |
| Self-reported health status at baseline Unknown or Not Reported | 16 Participants | 26 Participants | 10 Participants |
| Self-reported health status at baseline Very good | 87 Participants | 198 Participants | 111 Participants |
| Sex: Female, Male Female | 206 Participants | 404 Participants | 198 Participants |
| Sex: Female, Male Male | 138 Participants | 284 Participants | 146 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 364 | 0 / 364 |
| other Total, other adverse events | 0 / 364 | 0 / 364 |
| serious Total, serious adverse events | 0 / 364 | 0 / 364 |
Outcome results
Primary
Colorectal Cancer (CRC) Screening Completion
Completion of colonoscopy, fecal immunochemical testing (FIT), or other CRC screening test within 6 months of enrollment, based on documentation in the participants' electronic health record. If a patient completed both screening tests and the colonoscopy was a follow-up to a positive FIT, they were considered having completed a FIT only. If the FIT was negative and the patient still had a colonoscopy, they were considered to have completed both tests.
Time frame: 6 months post intervention
Population: All participants who agreed to having their electronic health record (EHR) checked by the researchers. There were 12 participants in the Verbal arm and 4 participants in the Quantitative arm who would not agree for their EHRs to be checked by the researchers; therefore, assessment of uptake was not performed with those participants.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Verbal | Colorectal Cancer (CRC) Screening Completion | 99 Participants |
| Quantitative | Colorectal Cancer (CRC) Screening Completion | 99 Participants |
Primary
Colorectal Cancer (CRC) Screening Intention: Change From Baseline to Post-intervention
Multiple choice question assessing subject's intention in getting a CRC screening test in the next 6 months. The response options were: 5=Definitely, 4=Probably, 3=May or may not, 2=Probably not, and 1=Definitely not. Higher positive change means CRC screening intention increased. Increased intention to be screened is a better outcome.
Time frame: 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid])
Population: All participants with both baseline and post-intervention responses to CRC screening intent items. There were 2 participants in each arm who did not respond to at least one of the intent items.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Verbal | Colorectal Cancer (CRC) Screening Intention: Change From Baseline to Post-intervention | 0.37 units on a scale | Standard Deviation 0.84 |
| Quantitative | Colorectal Cancer (CRC) Screening Intention: Change From Baseline to Post-intervention | 0.47 units on a scale | Standard Deviation 0.84 |
Secondary
Decision Conflict: Change in Conflict From Baseline to Post-intervention
Decision conflict is assessed using the16-item Decision Conflict Scale. Scores will range from 0 (low decision conflict) to 100 (high decision conflict). Negative change means decision conflict decreased. A decrease in decision conflict is a better outcome.
Time frame: 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid])
Population: All participants with at least 11 responses to the 16-item Decision Conflict Scale at both baseline (T0) and post-intervention (T1).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Verbal | Decision Conflict: Change in Conflict From Baseline to Post-intervention | -19.3 units on a scale | Standard Deviation 17.77 |
| Quantitative | Decision Conflict: Change in Conflict From Baseline to Post-intervention | -22.0 units on a scale | Standard Deviation 19.25 |
Secondary
Knowledge of Colorectal Cancer (CRC) and Colorectal Cancer Screening: Change in Number of Correct Answers From Baseline to Post-intervention
Qualitative knowledge is assessed with five multiple choice and five true/false questions regarding general information (including risk factors, screening test options, and test frequency) of CRC and CRC screening. Knowledge scores were derived by summing correct responses to the 10 individual knowledge questions (range, 0-10), and change from baseline was calculated using the knowledge score at post-intervention minus the knowledge score at baseline. The range for change could be 0 to 10 with higher values meaning an increase in knowledge which is a better outcome.
Time frame: 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid])
Population: All participants who completed at least 7 knowledge questions at both baseline (T0) and post-intervention (T1).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Verbal | Knowledge of Colorectal Cancer (CRC) and Colorectal Cancer Screening: Change in Number of Correct Answers From Baseline to Post-intervention | 3.13 correct answers | Standard Deviation 2.99 |
| Quantitative | Knowledge of Colorectal Cancer (CRC) and Colorectal Cancer Screening: Change in Number of Correct Answers From Baseline to Post-intervention | 3.66 correct answers | Standard Deviation 2.82 |
Secondary
Number of Patients With High and Low Subjective Numeracy
Subjective numeracy was determined using the Subjective Numeracy Scale (SNS), a validated instrument that involves 8 Likert-type questions regarding the participant's preference for or dislike of numeric information and their perception of their ability to understand it. Each item has a six-point response option from 1-6. The higher the number, the higher the subjective numeracy. All participants were divided into two groups: above (high numeracy) and below (low numeracy) the median for their total subjective numeracy score.
Time frame: 1 day
Population: Subjects who completed the Subjective Numeracy Scale.
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Verbal | Number of Patients With High and Low Subjective Numeracy | Low Subjective Numeracy | 172 Participants |
| Verbal | Number of Patients With High and Low Subjective Numeracy | High Subjective Numercy | 164 Participants |
| Quantitative | Number of Patients With High and Low Subjective Numeracy | Low Subjective Numeracy | 160 Participants |
| Quantitative | Number of Patients With High and Low Subjective Numeracy | High Subjective Numercy | 177 Participants |
Secondary
Perceived Barriers to Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) or Colonoscopy: Change From Baseline to Post-intervention
Assessed separately for colonoscopy (11 questions) and FIT (9 questions), using items drawn from validated scales for measuring barriers and self-efficacy of colonoscopy and FIT. All items had Likert-type response options where 5=strongly agree to 1=strongly disagree. Mean values were calculated within each set of questions (each combined score has a range of 1 to 5) at baseline and post-intervention and then change from baseline was calculated using the mean at post-intervention minus the mean at baseline. The range for change could be -4 to 4 with lower values meaning perceived barriers decreased which is a better outcome.
Time frame: 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid])
Population: Participants who completed both the baseline and post-intervention barrier questions. For colonoscopy, it was 322 in Verbal and 330 in Quantitative. For FIT, it was 321 in Verbal and 327 in Quantitative.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Verbal | Perceived Barriers to Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) or Colonoscopy: Change From Baseline to Post-intervention | Colonoscopy Barriers | -0.18 units on a scale | Standard Deviation 0.47 |
| Verbal | Perceived Barriers to Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) or Colonoscopy: Change From Baseline to Post-intervention | FIT Barriers | -0.18 units on a scale | Standard Deviation 0.55 |
| Quantitative | Perceived Barriers to Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) or Colonoscopy: Change From Baseline to Post-intervention | Colonoscopy Barriers | -0.13 units on a scale | Standard Deviation 0.45 |
| Quantitative | Perceived Barriers to Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) or Colonoscopy: Change From Baseline to Post-intervention | FIT Barriers | -0.21 units on a scale | Standard Deviation 0.54 |
Secondary
Perceived Benefits of Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) and Colonoscopy: Change From Baseline to Post-intervention
Assessed separately for colonoscopy (4 questions) and FIT (4 questions), using items drawn from validated scales for measuring benefits and self-efficacy of colonoscopy and FIT. All items had Likert-type response options where 5=strongly agree to 1=strongly disagree. Mean values were calculated within each set of questions (each combined score has a range of 1 to 5) at baseline and post-intervention and then change from baseline was calculated using the mean at post-intervention minus the mean at baseline. The range for change could be -4 to 4 with higher values meaning perceived benefits increased which is a better outcome.
Time frame: 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid])
Population: Participants who completed both the baseline and post-intervention benefit questions. For colonoscopy, it was 324 in Verbal and 332 in Quantitative. For FIT, it was 320 in Verbal and 330 in Quantitative.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Verbal | Perceived Benefits of Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) and Colonoscopy: Change From Baseline to Post-intervention | Colonoscopy Benefits | 0.44 units on a scale | Standard Deviation 0.62 |
| Verbal | Perceived Benefits of Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) and Colonoscopy: Change From Baseline to Post-intervention | FIT Benefits | 0.44 units on a scale | Standard Deviation 0.76 |
| Quantitative | Perceived Benefits of Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) and Colonoscopy: Change From Baseline to Post-intervention | Colonoscopy Benefits | 0.47 units on a scale | Standard Deviation 0.62 |
| Quantitative | Perceived Benefits of Colorectal Cancer Screening With Fecal Immunochemical Test (FIT) and Colonoscopy: Change From Baseline to Post-intervention | FIT Benefits | 0.52 units on a scale | Standard Deviation 0.73 |
Secondary
Perceived Risk of Colorectal Cancer (CRC): Change From Baseline to Post-intervention
Multiple choice questions assessing subject's perception of how likely they are to get colon cancer in the next 5 years, in the next 10 years, and sometime during their lifetime. Each had response options: 4=very likely, 3=somewhat likely, 2=somewhat unlikely, and 1=very unlikely. The mean of the 3 questions was calculated (each combined score has a range of 1 to 4) at baseline and post-intervention and then change from baseline was calculated using the mean at post-intervention minus the mean at baseline. The range for change could be from -3 to 3 with higher values meaning an increase in perceived risk which is a better outcome.
Time frame: 1 day (baseline [before viewing decision aid] and post-intervention [immediately after viewing decision aid])
Population: All participants with baseline (T0) and post-intervention (T1) perceived risk items. Missing responses were not used.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Verbal | Perceived Risk of Colorectal Cancer (CRC): Change From Baseline to Post-intervention | -0.08 units on a scale | Standard Deviation 0.7 |
| Quantitative | Perceived Risk of Colorectal Cancer (CRC): Change From Baseline to Post-intervention | 0.08 units on a scale | Standard Deviation 0.81 |