Breast and Prostate Cancer Data Quality and Patterns of Care Study in NC

描述（申请人提供）：背景：癌症登记传统上依赖于医院数据来估计人群中特定部位的癌症发病率。 在过去的几年里，疾病控制和预防中心（CDC）帮助扩大了登记处的作用，以提供有关癌症护理模式的关键信息。 因此，登记数据具有很大的潜力，可用于研究人口中癌症护理的趋势以及护理模式和弱势群体（如少数群体和服务不足者）在获得护理和结果方面的差异。 随着许多癌症的诊断和管理从传统的医院环境转移到门诊环境，需要来自医生办公室和其他非医院环境的更完整的诊断，治疗和随访信息。 我们已经发现，低数据质量的实例不是随机模式的，因此可以以已知的准确度进行统计预测。在此基础上，它是可行的，以制定有针对性的审计协议（TAP），达到或超过数据质量的程序阈值。 目的：这个为期3年的项目旨在提高北卡罗来纳州登记处最近第一疗程和分期数据的完整性、及时性、质量和使用;并描述在600例白色和非白色女性乳腺癌病例I-III期（N= 1，200）和600例患有前列腺癌的白色和非白色病例的随机选择样本中的护理模式，所有阶段（N=1，200）在NC注册中包含在最近两个诊断年可用。 这项工作建立在维克森林大学和北卡罗来纳州中央癌症登记处（NC CCR）过去关于癌症护理模式的合作工作基础上。在我们提出的工作中，我们计划实施有效的数据质量评估和改进策略，并按种族/民族群体和贫困标志物评估总样本中乳腺癌和前列腺癌的癌症模式。 具体而言，在这项为期三年的研究中，我们将：1）测试登记数据的目标审计协议（TAP）的效率和有效性，以提高登记数据的整体质量; 2）通过电子编辑和专业病例记录重新提取，评估NC CCR收集的分期和第一疗程的质量、完整性; 3）描述在北卡罗来纳州接受标准护理的女性乳腺癌报告病例的百分比; 4）描述北卡罗来纳州前列腺癌的护理模式，考虑患者（到辐射设施的距离、年龄、种族、疾病状态，如阶段/等级/血清标志物、城市/农村邮政编码）和系统或设施水平变量（手术量、注册医院、提供者类型）;以及5）参与汇总数据的分析，并遵循协作性研究小组的共同协议。

DESCRIPTION (provided by applicant): Background: Cancer registries traditionally have relied upon hospital data to estimate site-specific cancer incidence in a population. In the last few years, the Centers for Disease Control and Prevention (CDC) has helped expand the role of registries to provide key information on patterns of cancer care. As a result, registry data have great potential for research on trends of cancer care in the population as well as patterns of care and disparities in care access and outcomes of vulnerable groups such as minorities and the underserved. With the diagnosis and management of many cancers shifting from the traditional hospital setting to the ambulatory setting, more complete diagnostic, treatment, and follow-up information from physician offices and other non-hospital settings is required. We have found that instances of low data quality are not randomly patterned, and thus can be statistically predicted with known accuracy. On this basis it is feasible to develop targeted audit protocols (TAP) that meet or exceed program thresholds for data quality. Objectives: This 3-year project is designed to improve the completeness, timeliness, quality, and use of recent first course of treatment and stage data in the North Carolina registry; and to describe patterns of care in randomly selected samples of 600 white and nonwhite cases of female breast cancer cases stages l-lll (N=1,200) and 600 white and non-white cases with prostate cancer, all stages (N=1,200) contained in the NC registry in the two most recent diagnosis years available. This work builds upon the past collaborative work of Wake Forest University and the North Carolina Central Cancer Registry (NC CCR) on patterns of cancer care. In our proposed work, we plan to implement efficient strategies for data quality assessment and improvement, and assessment of patterns of cancer for breast cancer and prostate cancers within the total sample, by race/ethnicity groups, and by markers for poverty. Specifically, in this three year study we would: 1) Test the efficiency and effectiveness of a targeted audit protocol (TAP) for registry data to improve the overall quality of registry data; 2) Assess the quality, completeness of staging and first course of treatment collected by the NC CCR by electronic edits and professional case record re-abstraction; 3) Describe the percentage of reported cases with female breast cancer that receive standard of care in North Carolina; 4) Describe the patterns of care with prostate cancer in North Carolina, considering patient (distance to radiation facility, age, race, disease status such as stage/grade/serum markers, urban/rural zip code) and system, or facility level variables (procedure volume, registry hospital, provider type); and 5) Participate in the analysis of aggregate data and follow common protocols of the collaborative PoC group.