Risk Stratification to Promote Effective Shared Decision-Making for Colorectal Cancer Screening

Shared decision-making (SDM) has been advocated as a strategy for increasing colorectal cancer (CRC) screening rates. Our studies to date suggest that while the use of a novel computer-based decision aid facilitates several components of SDM from both the patient and provider perspective, there is a reluctance among providers to acquiesce to patient preferences for a particular screening strategy when its differs from their own. The overall objective of this study is to assess whether risk stratification for advanced colorectal neoplasia influences clinical decision-making related to screening test selection and adherence within a SDM framework. Eligible subjects will be randomized to either an experimental arm, in which they will be asked to complete a 6-item risk assessment questionnaire known as the "Advanced Colorectal Neoplasia Index [ACNI]" after reviewing a web-based decision aid, or a control arm, in which they will only review the decision aid. Both interventions will take place just before a prearranged office visit with their provider. The primary outcome will be screening test ordered; secondary outcomes will include test completion rates, concordance between test preference and test ordered,, patient satisfaction with decision-making process, screening intentions, 6-month test completion rates and provider satisfaction. Outcomes will be evaluated using computerized tracking systems or validated instruments.

Colorectal cancer (CRC) is the second leading cause of cancer-related death in the United States. Screening by any of at least 6 different methods is a cost-effective yet underutilized strategy for reducing both CRC incidence and mortality. Because these methods differ with respect to risks and benefits and because existing evidence fails to identify a single best strategy, most authoritative groups advocate a shared decision-making (SDM) approach when selecting an appropriate screening strategy. SDM is a sequential, interactive process involving information exchange, values clarification, decision-making and mutual agreement. To facilitate this process, patient-oriented decision aids have been developed to enable patients to identify a preferred strategy based on personal values and empower them to participate in the decision-making process. Our recent studies to date find that although decision aids enable patients to make informed choices, providers are often unwilling to acquiesce to patient preferences when they differ from their own. Since accurate risk assessment is a critical component of effective clinical decision-making, the investigators postulate that risk stratification for the point prevalence of advanced colorectal neoplasia will enable providers to incorporate objective risk-based criteria in their decision-making when considering patient preferences for screening. To that end, the investigators have recently developed and validated the so-called "Advanced Colorectal Neoplasia Index [ACNI]" that stratifies patients into low versus intermediate/high risk categories based on available clinical data, including age, sex, race/ethnicity, smoking history, daily alcohol intake and use of non-steroidal anti-inflammatory drugs. The overall objective of this study is to determine whether risk stratification using the ACNI influences clinical decision-making related to screening test selection and adherence to screening within a SDM framework. Hypothesis: Providers who incorporate risk estimates of ACN in their decision-making when recommending screening tests are more likely to consider patient preferences for options other than colonoscopy than providers lacking this information.

Colorectal cancer screening, Shared decision-making, Patient preferences, Risk assessment, Advanced colorectal neoplasia

Subjects randomized to the control arm will review the web-based decision aid (http://www.colorectalcancerscreening4u.com) just prior to a scheduled visit with their provider.

Subjects randomized to the experimental arm will complete the ACNI risk assessment tool after reviewing the web-based decision aid (http://www.colorectalcancerscreening4u.com) just prior to a scheduled office visit with their provider.

Patients randomized to the experimental arm will be asked a complete the ACNI risk assessment tool after reviewing a web-based colorectal cancer decision aid. The ACNI uses a point based system to stratify patients into low (mean rate of ACN ~3%) versus intermediate/high (~ 8%) risk groups based on responses to 6 items: age (50-59, 60-69, 70+), sex (male/female), race/ethnicity (non-Hispanic black, other), smoking history (never, <20 years, 20+ years), daily alcohol intake (< 2 vs. >/=2 drinks) and use of non-steroidal anti-inflammatory drugs (ever, never). The index represents a prototype version of the Advanced Colorectal Neoplasia Index (Am J Gastroenterol 2015;110:1062-71).

Concordance is a measure of the agreement between the patient's test preference and actual test ordered for standard care vs. risk assessment patients. It is defined as the number of patients who had their preferred test ordered.

Concordance between patient preference and test ordered for high versus low risk patients. It is defined as the number of patients who had their preferred test ordered.

SDMP was assessed on the posttest using the validated 12-item Satisfaction with the Decision-Making Process scale. Individual items are assigned a point value ranging from 1 for ''strongly disagree'' (or ''poor'') to 5 for ''strongly agree'' (or ''excellent''). A cumulative score is then calculated based on the summed response scores for each item (maximum score = 60). Data was missing for 11 patients in the concordant group and 6 patients in the discordant group

Screening intentions were assessed on the posttest. Patients were asked how sure they were that they would complete the screening test that got scheduled Scores ranged from 5 = ''completely'' to 1 = ''not at all sure.'' Data was missing for 11 patients in the concordant group and 6 patients in the discordant group.

Test completion rates were tracked using BMC's electronic medical record, which captures results for all endoscopic procedures, imaging studies, and stool blood tests.

Provider satisfaction was assessed based on responses to a 3-item pretest administered prior to commencement of the study and the same 3-item posttest. The 3 items assessed to the extent to which providers felt that personalized risk assessment would be useful for: (1) selecting an appropriate screening test for their average risk patients [test selection]; (2) reduce time to decide on an appropriate screening modality [save time]; and (3) make them more receptive to patient preferences and possibly order a screening test other than colonoscopy [receptive to patient preferences]. Responses were assigned a point value ranging from 5= "strongly agree" and 1 = "strongly disagree".

Test-specific concordance between patient preference for colonoscopy and test ordered for standard care versus risk assessment groups. It is defined as the number of patients who had their preferred test ordered.

Test-specific concordance between patient preference for a screening test other than colonoscopy (fecal occult blood testing, flexible sigmoidoscopy, double-contrast barium enema, CT colonography and stool DNA) and test ordered for standard care versus risk assessment arms. It is defined as the number of patients who had their preferred test ordered.

Concordance Between Patient Preference for Colonoscopy and Test Ordered for High Versus Low Risk Patients

Test-specific concordance between patient preference for colonoscopy and test ordered for high versus low risk patients. It is defined as the number of patients who had their preferred test ordered.

Concordance Between Patient Preferences for a Screening Tests Other Than Colonoscopy and Test Ordered for High Versus Low Risk Patients

Test-specific concordance between patient preference for a screening test other than colonoscopy (fecal occult blood testing, flexible sigmoidoscopy, double-contrast barium enema, CT colonography and stool DNA) and test ordered for high versus low risk patients. It is defined as the number of patients who had their preferred test ordered.

Inclusion Criteria: English-speaking "average-risk" patients 50 to 75 years of age; Due for CRC screening based on current recommendations (i.e. no prior screening or > 1year since last fecal occult blood testing [FOBT], > 3 years since last stool DNA test, > 5 years since last flexible sigmoidoscopy, virtual colonoscopy or double-contrast barium enema [DCBE], or > 10 years since last colonoscopy); Under the direct care of a staff (attending) primary care provider or physician extender; Absence of major co-morbidities that preclude CRC screening. Exclusion Criteria: High-risk condition (personal history of colorectal cancer or polyps, family history of colorectal cancer or polyps involving one or more first degree relatives < 60 years of age, chronic inflammatory bowel disease); Presence of "alarm" gastrointestinal symptoms, including rectal bleeding, recent change in bowel habits, abdominal pain, unexplained weight loss and iron deficiency anemia; Comorbidities that preclude CRC screening by any method; Lack of fluency in written and spoken English (since decision aid and personalized risk assessment tool will be in English only due to funding issues).

Schroy PC 3rd, Emmons K, Peters E, Glick JT, Robinson PA, Lydotes MA, Mylvanaman S, Evans S, Chaisson C, Pignone M, Prout M, Davidson P, Heeren TC. The impact of a novel computer-based decision aid on shared decision making for colorectal cancer screening: a randomized trial. Med Decis Making. 2011 Jan-Feb;31(1):93-107. doi: 10.1177/0272989X10369007. Epub 2010 May 18.

Schroy PC 3rd, Mylvaganam S, Davidson P. Provider perspectives on the utility of a colorectal cancer screening decision aid for facilitating shared decision making. Health Expect. 2014 Feb;17(1):27-35. doi: 10.1111/j.1369-7625.2011.00730.x. Epub 2011 Sep 8.

Schroy PC 3rd, Caron SE, Sherman BJ, Heeren TC, Battaglia TA. Risk assessment and clinical decision making for colorectal cancer screening. Health Expect. 2015 Oct;18(5):1327-38. doi: 10.1111/hex.12110. Epub 2013 Jul 30.

Schroy PC 3rd, Wong JB, O'Brien MJ, Chen CA, Griffith JL. A Risk Prediction Index for Advanced Colorectal Neoplasia at Screening Colonoscopy. Am J Gastroenterol. 2015 Jul;110(7):1062-71. doi: 10.1038/ajg.2015.146. Epub 2015 May 26.

Schroy PC 3rd, Duhovic E, Chen CA, Heeren TC, Lopez W, Apodaca DL, Wong JB. Risk Stratification and Shared Decision Making for Colorectal Cancer Screening: A Randomized Controlled Trial. Med Decis Making. 2016 May;36(4):526-35. doi: 10.1177/0272989X15625622. Epub 2016 Jan 19.