Cancer Risk Assessment Tools in Primary Care: A Systematic Review of Randomized Controlled Trials

Based on their systematic reviews of randomized clinical trials, Dr. Walker and her colleagues conclude that use of existing cancer risk assessment tools in the primary care setting increase screening intentions but their impact on screening behavior is less certain. This conclusion was based in part on their critique of our work (1,2), which found that patients randomized to a decision aid plus personalized risk assessment arm expressed higher screening intentions than controls but no more likely to complete a screening test within a 12-month time frame. Perhaps even noteworthy, however, was the observation that these individuals were actually less likely than those in the decision aid alone intervention arm to complete a screening test within this same time frame.

Several factors might have contributed to the negative association between personalized risk feedback and screening behavior. The lack of appropriate framing when conveying risk status k might have deflated inflated interest in screening for both those deemed to be "above" and "bellow" average risk. Individuals deemed to be at higher risk might have been fearful of potential findings and thus more motivated from gain-frame messaging promoting the benefits of screening; conversely those at lower risk might have felt that screening was unnecessary and thus more motivated from loss-framed messaging regarding the "costs" of not screening. The choice of endpoint by which to define risk might also have been a deterrent. Estimates of future cancer risk derived using the YourDiseaseRisk tool might have been less motivating than a real-time estimates of risk for existing advanced colorectal neoplasia (advanced adenomas and presymptomatic cancers), especially among those to be deemed at increased risk. Lastly, the lack of clinician engagement might have also been a mitigating factor. Sharing risk estimates with the patient's primary care clinician during the study visit might have facilitated more effective risk communication and an objective basis for selecting a mutually-agreeable screening test in situations where clinician and patient preferences differ. Hopefully, these considerations will inform the design of future studies aimed at exploring the impact of risk assessment on colorectal cancer screening adherence.

REFERENCES

1. Schroy P, Emmons K, Peters E, Glick J, Robinson P, Lydotes M, Mylvaganam S, Evans S, Chaisson C, Pignone M, Prout M, Davidson P, Heeren T. The impact of a novel computer-based decision aid on shared decision- making for colorectal cancer screening: A randomized trial. Med Decis Making. 2011;31:93-107

2. Schroy P, Emmons K, Peters E, Glick J, Robinson P, Lydotes M, Mylvaganam S, Coe A, Chen C, Chaisson C, Pignone M, Prout M, Davidson P, Heeren T. Aid-assisted decision making and colorectal cancer screening: a randomized controlled trial. Am J Prev Med. 2012;43:573-83.

Competing interests: None declared