1A
ORAL ABSTRACTS: PREFERENCES AND DECISION MAKING IN COLORECTAL CANCER SCREENING
Method: We conducted a single-blinded, randomized controlled trial to test the efficacy of a CRC DA designed for older adults. Eligible patients were asked to arrive early for an upcoming clinic visit to receive the DA or control materials. Before seeing their provider, participants completed written questions including: likelihood of discussing screening, physician’s screening preference, preferences for decision making process, life expectancy, likelihood that screening will prolong life, and decisional balance between risks and benefits.
Result: 424 participants ages 70-84 were recruited from 14 primary care practices within the Duke Primary Care Research Consortium. The intervention group was more likely than the control to say they were very/somewhat likely to talk with their doctor about CRC screening (60% vs 44%; p=.001), versus somewhat/very unlikely. In both groups, >78% of participants preferred to share the decision making process with their doctor in some capacity. A greater proportion of patients in the intervention group than the control thought they would live >10 years (47% vs. 39%; p=.12) versus less time, although it was not statistically significant. Patients who received the DA were less likely than the control to believe screening would prolong their lives “a lot” (12% vs. 23%; p=.005), compared to any shorter duration of time. However, >40% of all participants were “unsure” whether screening would prolong their lives. The intervention group was less likely than the control to say the “Benefits greatly outweigh the risks” of screening (30% vs. 46%; p=.001), compared to any other category. There was no significant difference between the groups’ perceptions of their providers’ screening preference.
Conclusion: Exposure to the DA appears to increase patients’ intent to discuss screening with providers. The DA also appeared to affect risk/benefit assessment, perhaps due to increased knowledge about screening. It is interesting that a plurality of respondents in both groups remain unsure about whether screening would prolong their lives. Although not statistically significant, differences in perceived life expectancy may be clinically important, as the data suggest a trend to report a higher life expectancy among intervention patients.
Method: We performed confirmatory factor analysis, regression analysis, mediation analysis, and Classification and Regression Trees analysis on baseline patient survey data from a 2-armed randomized controlled study (R01CA152413) set in 12 community and 3 university-based primary care practices in Metro Detroit. Patients were men and women aged 50 to 75 years not current on CRC screening. Main outcomes were the patient intent to get screened for CRC and patient preference for a particular CRC screening test. Perceived risk and self-efficacy were intermediate outcomes.
Result: Data were obtained from 570 participants. Mean age was 57.7 years, 56.1% were women, and 55.1% were white and 36.6% black. Women had 32% lower odds than men to perceive CRC to be high/moderate risk [OR 0.68 (0.47-0.97), p=0.03], and 41% lower odds than men of having high self-efficacy [OR 0.59 (0.42-0.85), p=0.006]. Whites had 63% lower odds than blacks of having high self-efficacy [OR 0.37 (0.25-0.57), p<0.001], and 47% lower odds to have intent to get screened for CRC [OR 0.53 (0.34-0.84), p=0.007]. Older age, higher knowledge, lower level of test worries, and medium/high vs. low self-efficacy increased the odds of being intent on getting screened for CRC. Self-efficacy, but not perceived risk, significantly mediated the association between race, attitude, and test worries and patient intent to get screened for CRC. Neither self-efficacy nor perceived risk significantly mediated for CRC screening test preference. However, participants (n=30) who wanted neither stool blood test nor colonoscopy had about 20% lower odds of being intent on getting screened for CRC than the ones with specific [OR 0.82 (0.72-0.94), p=0.004] or non-specific [OR 0.77 (0.67- 0.89), p<0.001] test preference.
Conclusion: Age, gender, and race were significantly associated with perceived risk, self-efficacy, and intent. Self- efficacy significantly mediated the association between race, attitude, and test worries and patient intent to get screened. Having neither preference for stool blood test nor colonoscopy negatively correlated with intent.
Method: A single-blinded, randomized controlled trial testing the effectiveness of a written decision support intervention targeted for older adults making a decision about whether to undergo CRC screening compared to an attention control about driving safety. Participants were recruited from 14 primary care practices affiliated with the Duke Primary Care Research Consortium. Eligible patients were ages 70 to 84, had upcoming appointments within 4 to 6 weeks, and were eligible for CRC screening. The primary outcome was appropriate CRC screening behavior six months after the index visit. Appropriate screening behavior was defined as screening test completion for participants in the good health state, discussion about CRC screening in the intermediate health state (preference sensitive), and no screening test completion in the poor health state. Health state classifications were derived from age and Charlson Comorbidity Index and were based on the life expectancy needed to expect a net benefit from CRC screening.
Result: Of 424 patients randomized, 412 (97%) had complete data for analysis. The proportion of participants who exhibited appropriate CRC screening behavior at six months was higher in the decision support group compared to control (55% vs. 45%; p=0.03). Consistent with our hypothesis, a higher proportion of participants in the good health state were screened in the intervention group compared to control group (27% vs 13%; p=0.01) and in the intermediate health state, a higher proportion of participants in the intervention group reported discussions compared to controls. (53% vs 38%; p=0.02). We did not find a difference in the proportion who did not undergo screening among participants the poor health state (92% PDA vs 95% control; p=0.56).
Conclusion: Overall, the decision aid promoted appropriate CRC screening, primarily in participants with good or intermediate health status. Low rates of screening in participants with poor health status limited our ability to test the effectiveness of the decision aid in decreasing overuse in this subgroup. Targeted patient decision aids may be an effective method to promote appropriate care in an older population along the continuum of health.
Title: Influences on Primary Care Providers' Recommendations for Hypothetical Prostate and Colon Cancer Screening in Older Adults: a Within-subjects Experiment
Abstract (word count max 375)- currently 283
Purpose: To identify factors that influence primary care providers' recommendations for cancer screening in older adults.
Methods: A cross-sectional survey of all primary care providers (n=123) in a Research Consortium of 24 primary care Family Medicine and General Internal Medicine clinics in central North Carolina. We designed a within-subjects experiment in which providers read two hypothetical patient vignettes that varied by type of cancer screening test (prostate and colorectal) and were asked whether they would recommend screening for patients that varied by age (age 50, 70, 90) and request for screening (request and no request). Outcomes were providers' recommendations for cancer screening.
Results: We found 51% (95% CI 48%, 54%) would recommend cancer screening, regardless of screening test type, patient age, or patient request. Providers reported they would recommend screening more often for colorectal cancer as compared to prostate cancer (p<0.001) and for younger patients as compared to older patients (p>0.001) Providers said they would recommend screening more often when the patient requested it as compared to when patients made no request for screening, for both vignettes (prostate and colorectal cancer screening) across all ages (p<0.001). The recommendation for screening of hypothetical younger patients was more pronounced for colorectal cancer screening as compared to prostate cancer screening (p<0.001 for the interaction of age on the relationship between screening test and recommendation). The interaction of age on the relationship between patient request and recommendation approached statistical significance (p=0.08).
Conclusions: Providers' reported recommendations for cancer screening depended on type of hypothetical patient being screened. Colorectal cancer screening, younger age, and patient request all increased providers' recommendations. Next steps may include the development of interventions to address screening for prostate cancer, older adults, or patient request.
Methods: 622 participants aged 50-75 were recruited from a national on-line survey panel. All participants viewed information about CRC and two approved tests. Participants were then randomized to view additional information following a 2x3 between-subjects experimental design (Table 1):
(a) one of three personalized-risk conditions: no personalized risk information, or a hypothetical scenario in which lifetime risk of CRC mortality was either 1.5% (low personalized risk) or 6% (high personalized risk).
(b) one of two average-risk conditions: viewing vs. not viewing average lifetime risk of CRC mortality (3%). Viewing average risk allowed individuals to determine whether their personalized risk was above or below average (“comparative risk”).
The no-risk-information condition served as the control group. Participants then completed a questionnaire assessing perceived CRC risk, screening intention and test choice.
Results: There was a significant interactive effect of average and personalized risk information on perceived likelihood of dying of CRC (F2,622 = 4.40, p=.013, eta2=.014).
Participants who viewed only average-person risk information, and participants assigned to the low personalized risk condition, both had significantly lower perceived CRC mortality risk than participants in the no-risk-information control group. However, participants assigned to the high personalized-risk condition did not have significantly different perceived CRC risk than the control group. (Table 1.)
There was no significant difference between experimental groups in intent to be screened or in test choice. Numeracy did not moderate the effects of the experimental conditions on perceived risk, intent, or test choice.
Conclusions: Low personalized risk information decreased participants’ perceived CRC mortality risk; however, high personalized risk information had no impact on individuals’ perceived CRC risk. Neither personalized nor average risk information influenced participants’ screening intentions. These findings suggest that personalized and comparative risk data may only influence behavior if that data is supplemented with description of potential implications for screening.
Method: A web-based survey was developed for persons aged 50-75 who were at average risk of CRC, but screening naive. Respondents were asked a series of questions to gauge their perspectives on CRC screening and knowledge of five existing CRC screening tests (colonoscopy, flexible sigmoidoscopy (FS), fecal occult blood testing (FOBT), fecal immunochemical testing (FIT), and stool DNA testing). Respondents were then introduced to a series of CRC test profiles, each of which contained a description, as well as information on dietary preparation, time requirements, physical discomfort, complication risk, frequency of testing, accuracy, and follow up of abnormal results. After reviewing the test profiles, respondents were asked another series of questions to elicit their preferred screening option and the CRC test attributes influencing their choice.
Result: To date, 415 eligible persons have completed the survey. Among the 83% of respondents indicating their likelihood of undergoing screening in the next year was a 3 or higher on a 5-point Likert-type scale (1=definitely not, 3=maybe, 5=definitely) after being introduced to the test profiles, 40% would choose to receive colonoscopy and 37% would choose the new stool DNA test. Few respondents indicated they would choose FOBT (10%), FOBT (9%), or FS (4%). Among persons choosing colonoscopy, accuracy of the test at detecting cancer and polyps and frequency of testing were cited as the main drivers of preference (64%, 63%, and 41% of respondents, respectively). Among respondents choosing stool DNA, the low level of discomfort associated with the test, amount of time required to do the test, and test process were cited as the primary drivers of preference (45%, 41%, and 25% of respondents, respectively).
Conclusion: Educating patients about the full menu of recommended screening options, including stool DNA testing, and eliciting patient preferences that reflect the importance they place on individual test features can potentially increase participation among the unscreened population.