A clinical prediction rule for identifying South African colorectal cancer patients who will fail to engage in oncology care

用于识别无法参与肿瘤护理的南非结直肠癌患者的临床预测规则

• 批准号: 10689780
• 负责人: Yoshan Moodley
• 金额: $ 6.08万
• 依托单位: UNIVERSITY OF KWAZULU-NATAL
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-02 至 2026-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10689780
• 关键词: Address; Adherence; Advisory Committees; Algorithmic Analysis; Applications Grants; Appointment; Artificial Intelligence; Award; Biopsy; Books; Caring; Characteristics; Clinic; Clinic Visits; Clinical; Colorectal Cancer; Communities; Competence; Country; Coupled; Data; Data Collection; Data Sources; Decision Making; Diagnosis; Disease; Disease Management; Electronics; Enrollment; Epidemiologist; Failure; Funding; Future; Government; Grant; Health Care Sector; Healthcare Systems; Intention; Intervention; Laboratories; Literature; Logistic Regressions; Machine Learning; Malignant Neoplasms; Malignant neoplasm of gastrointestinal tract; Mentors; Meta-Analysis; Newly Diagnosed; Oncologist; Oncology; Operative Surgical Procedures; Outcome; Outcome Study; Patient-Focused Outcomes; Patients; Population; Prevention strategy; Program Development; Public Health; Public Sector; Publishing; Questionnaires; Research; Research Assistant; Research Personnel; Resource-limited setting; Resources; Review Literature; South Africa; South African; Specialist; Statistical Data Interpretation; Surgical Oncologist; Surveys; Time; Training; Universities; Validation; Visit; Vocational Guidance; advanced disease; burden of illness; care systems; career development; clinical practice; clinical predictors; cohort; colon cancer patients; colorectal cancer screening; comorbidity; data registry; design; electronic data; electronic registry; feature selection; flexibility; gradient boosting; high risk; improved; machine learning algorithm; machine learning method; model development; mortality; neural network; novel; patient registry; predictive modeling; prospective; random forest; scale up; screening; screening program; sociodemographics; statistics; support vector machine; systematic review; tool

PROJECT SUMMARY/ABSTRACT Colorectal cancer (CRC) is an emerging public health problem in South Africa (SA). Preliminary research from KwaZulu-Natal (KZN), SA shows that 23% of newly diagnosed CRC patients in the public healthcare sector will fail to engage in oncology care (FEOC). This complicates oncologists’ and oncologic surgeons’ efforts to effectively manage CRC disease in this setting. There are a variety of resource-intensive interventions for addressing FEOC. However, implementing these in all newly diagnosed CRC patients would be unfeasible, given the high patient volumes and existing resource constraints in the SA public healthcare sector. A clinical prediction rule (CPR) is proposed for identifying newly diagnosed SA CRC patients who are at high risk of FEOC, thereby enabling more efficient targeting of these specific, high risk patients for interventions which seek to address FEOC. The CPR will be developed by: 1) Identifying potential predictors of FEOC in CRC populations through a systematic review of the literature and a survey of 109 patients; 2) Constructing an appropriate electronic data collection tool that will streamline prospective data collection for these potential predictors and the study outcome, FEOC, from a cohort of 398 newly diagnosed CRC patients; and 3) Applying machine learning methods to the data to derive and internally validate the CPR. The CPR will be an important first step in tackling the high FEOC rates amongst newly diagnosed CRC patients in KZN, SA and will ultimately contribute toward improving both CRC disease management and patient outcomes in this setting. The proposed K43 research has the potential to be scaled up to the national level, and this will be the basis of a future R01 grant application. Dr. Moodley (K43 applicant) wants to become an independent gastrointestinal cancer epidemiologist and a global leader in CPRs, particularly CPRs that seek to address breakdowns in linkage to care and reduce the burden of disease attributed to gastrointestinal cancers, such as CRC. This award will provide him with protected time to conduct the proposed research and engage in an intensive career development program while based at one of the top universities in SA, the University of KwaZulu-Natal. He will be mentored by esteemed senior researchers from SA (Prof. Thandinkosi Madiba, University of KwaZulu-Natal) and the US (Prof. Ravi Pokala Kiran, Columbia University) who have a strong track record of mentoring young researchers. A research advisory committee, comprised of other senior researchers, will provide additional research and career guidance to Dr. Moodley. This will be coupled with didactic and “hands on” training in the core competencies required to develop CPRs. It is envisioned that by the end of his K43 training, Dr. Moodley will have developed the proposed CPR and generated new preliminary data. He will use this new data to successfully apply for an R01 grant, thus establishing his independence as gastrointestinal cancer epidemiologist and a global leader in CPRs.

项目总结/摘要 结直肠癌（CRC）是南非（SA）的一个新兴公共卫生问题。初步研究从 南非夸祖鲁-纳塔尔（KZN）的研究表明，在公共医疗保健部门，23%的新诊断CRC患者将 未能参与肿瘤护理（FEOC）。这使得肿瘤学家和肿瘤外科医生的努力变得复杂， 在这种情况下有效地管理CRC疾病。有各种资源密集型干预措施， 向联邦平等机会委员会提出。然而，在所有新诊断的CRC患者中实施这些是不可行的， 考虑到SA公共医疗保健部门的高患者数量和现有资源限制。临床 预测规则（CPR）提出了用于识别新诊断的SA CRC患者谁是高风险的， FEOC，从而能够更有效地针对这些特定的高风险患者进行干预， 来解决FEOC的问题CPR将通过以下方式制定：1）确定CRC人群中FEOC的潜在预测因素 通过系统的文献回顾和对109例患者的调查; 2）构建一个合适的 电子数据收集工具，将简化这些潜在预测因素的预期数据收集， 来自398名新诊断的CRC患者的研究结果FEOC;以及3）应用机器 学习方法的数据，以获得和内部验证CPR。CPR将是重要的第一步 在解决KZN，SA新诊断的CRC患者中的高FEOC率，并最终将有助于 在这种情况下，改善CRC疾病管理和患者结局。建议的K43 研究有可能扩大到国家一级，这将是未来R 01赠款的基础 应用程序. Moodley博士（K43申请人）希望成为一名独立的胃肠道癌症流行病学家 和全面政策建议的全球领导者，特别是寻求解决与护理和减少 胃肠道癌症（如CRC）造成的疾病负担。这个奖项将为他提供 保护时间进行拟议的研究，并从事密集的职业发展计划，同时 总部设在南非的顶尖大学之一，夸祖鲁-纳塔尔大学。他将由一位受人尊敬的 来自南非（Thandinkosi Madiba教授，夸祖鲁-纳塔尔大学）和美国（Ravi教授， 哥伦比亚大学的Pokala Kiran），他们在指导年轻研究人员方面有着良好的记录。一个研究 由其他高级研究人员组成的咨询委员会将提供额外的研究和职业指导 穆德利医生这将与所需核心能力的教学和“动手”培训相结合， 制定CPR。据设想，在他的K43培训结束时，Moodley博士将开发出拟议的 CPR并生成新的初步数据。他将使用这些新数据成功申请R 01补助金， 确立了他作为胃肠道癌症流行病学家和CPR全球领导者的独立性。

项目成果

Self-adherence to post-colonoscopy consults in patients undergoing diagnostic colonoscopy: Findings from a cross-sectional, quantitative survey at a South African quaternary hospital.

• DOI: 10.1371/journal.pone.0288752
• 发表时间: 2023
• 期刊: PLOS ONE
• 影响因子: 3.7
• 作者: Moodley, Yoshan;van Wyk, Jacqueline;Ning, Yuming;Wexner, Steven;Gounden, Cathrine;Naidoo, Vasudevan;Kader, Shakeel;Neugut, Alfred I.;Kiran, Ravi P.
• 通讯作者: Kiran, Ravi P.