A Risk Prediction Model for Colorectal Cancer Screening

描述（由申请人提供）：共同决策（SDM）已被提倡作为提高结直肠癌（CRC）筛查率的策略。本研究的总体目标是评估晚期结直肠肿瘤风险预测模型的发展是否是促进有效SDM和提高结肠镜筛查成本效益的可行和有效策略。在目标1中，我们建议对4000例接受筛查性结肠镜检查的无症状患者进行横断面调查，目的是基于假定的可改变和不可改变的风险因素开发预测模型，将患者分为晚期结直肠肿瘤的不同风险类别。将要求受试者在检查前完成一份简短的风险评估问卷;回答将与内镜检查结果相关。一旦得出，我们将在种族/民族多样化的患者人群（n=1200）中评价模型的性能特征。目标#2是调整模型，以用作参与SDM的初级保健医生在与患者讨论CRC筛查时的决策支持工具。我们建议使用定性和定量的方法来检查影响供应商决策的因素，评估供应商的计算能力，并确定定义风险类别的阈值。一旦定义，我们将把该模型转化为一个基于网络的风险评估工具，能够生成患者衍生的风险估计与可接受的框架，优化风险沟通。将举办实施前研讨会，培训服务提供者使用该工具、病历中的适当记录以及向患者传达风险的战略。目的#3是进行临床试验，以确定风险分层是否影响SDM框架内与筛选试验选择相关的临床决策。合格受试者（n=440）将被随机分配完成风险评估工具或不完成风险评估工具，然后审查基于网络的决策辅助工具，该工具描述了5种推荐的CRC筛查选项的利弊。这两种干预措施都将在与其提供者预先安排的办公室访问之前进行。主要结局将是筛选测试顺序;次要结局将包括测试完成率、价值一致性、患者对决策过程的满意度和提供者满意度。将使用计算机化跟踪系统或经验证的仪器对结局进行评价。目的#4是确定与普遍使用筛查性结肠镜检查和其他目前推荐的筛查策略相比，基于风险使用筛查性结肠镜检查的成本效益。公共卫生相关性：共同决策（SDM）已被提倡作为提高结直肠癌（CRC）筛查率的策略。晚期结直肠肿瘤有效风险预测模型的开发将使供应商在推荐首选筛查方案时将风险估计纳入其决策，从而促进更有效的SDM并提高筛查结肠镜检查的成本效益。

DESCRIPTION (provided by applicant): Shared decision-making (SDM) has been advocated as strategy for increasing colorectal cancer (CRC) screening rates. The overall objective of this study is to assess whether the development of a risk prediction model for advanced colorectal neoplasia is a feasible and valid strategy for facilitating effective SDM and improving the cost-effectiveness of screening colonoscopy. In Aim #1, we propose to conduct a cross- sectional survey of 4000 asymptomatic patients undergoing screening colonoscopy for the purpose of developing a prediction model based on putative modifiable and non-modifiable risk factors that stratifies patients into distinct risk categories for advanced colorectal neoplasia. Subjects will be asked to complete a brief risk assessment questionnaire prior to their examination; responses will be correlated with endoscopic findings. Once derived, we will evaluate the model's performance characteristics among a racially/ethnically diverse patient population (n=1200). Aim #2 is to adapt the model for use as a decisional support tool for primary care physicians who engage in SDM when discussing CRC screening with their patients. We propose to use qualitative and quantitative methods to examine factors that influence provider decision-making, assess provider numeracy skills, and identify thresholds for defining risk categories. Once defined, we will translate the model into a web-based risk assessment tool capable of generating patient-derived risk estimates with acceptable framing that optimizes risk communication. Pre-implementation seminars will be conducted to train providers in use of the tool, appropriate documentation in the medical record and strategies for communicating risk to patients. Aim #3 is to conduct a clinical trial to determine whether risk stratification influences clinical decision-making related to screening test selection within a SDM framework. Eligible subjects (n=440) will be randomized to complete the risk assessment tool or not prior to reviewing a web-based decision aid describing the pros and cons of the five recommended CRC screening options. 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, value concordance, patient satisfaction with decision- making process and provider satisfaction. Outcomes will be evaluated using computerized tracking systems or validated instruments. Aim #4 is to determine the cost-effectiveness of risk-based use of screening colonoscopy compared to universal use of screening colonoscopy and other currently recommended screening strategies. PUBLIC HEALTH RELEVANCE: Shared decision-making (SDM) has been advocated as strategy for increasing colorectal cancer (CRC) screening rates. The development of a valid risk prediction model for advanced colorectal neoplasia would enable providers to incorporate risk estimates into their decision-making when recommending a preferred screening option, thus facilitating more effective SDM and improving the cost-effectiveness of screening colonoscopy.