CCW Chronic Conditions

CCW Chronic Condition Segments Overview CCW Chronic Condition Segments can be requested in addition to the Master Beneficiary Summary File (MBSF) base file. These segments provide standardized indicators of whether a beneficiary has specific chronic conditions, derived from Medicare claims data using CMS algorithms. Each record in the Chronic Condition segment links to the beneficiary in the MBSF base file through BENE_ID, allowing researchers to combine chronic condition flags with demographics, enrollment, and mortality data. Purpose and Methodology The CCW chronic condition indicators were created to simplify research by applying validated claims-based algorithms to identify common chronic diseases, removing the need for researchers to define their own coding logic. These algorithms scan inpatient, outpatient, carrier, and other claims files over a defined lookback period of one to three years, depending on the condition. To ensure reliability, beneficiaries must also have sufficient fee-for-service (FFS) coverage to capture all relevant claims; those enrolled in managed care (MA/HMO) plans often lack complete claims data. Versions Available The Chronic Conditions Data Warehouse (CCW) provides two versions of standardized chronic condition indicators: 1. 27 CCW Chronic Conditions Developed by CMS using algorithms based on validated research literature and federal guidelines. These indicators identify the presence of chronic conditions using Medicare claims data over a defined lookback period. Uses both ICD-9-CM and ICD-10-CM diagnosis codes, depending on the year. CMS Coverage: 1999–2021 2. 30 CCW Chronic Conditions In 2020, CMS collaborated with an expert panel to refine and enhance the original 27-condition algorithms. This led to the expanded and improved 30 CCW Chronic Conditions, which reflect updated clinical standards, enhanced claims logic, and improved identification accuracy. Algorithms are based entirely on ICD-10-CM diagnosis codes (no ICD-9). CMS Coverage: 2017-present Access and Usage These datasets are restricted-use due to the inclusion of sensitive and identifiable health information. As such, the data files themselves cannot be shared via this Dataverse entry. Students, post-docs, and researchers conducting NSAPH-related research in collaboration with PIs at Harvard or affiliated institutions may gain access to the data, provided they have submitted a project initiation form, are listed on an approved IRB protocol, and have completed all required data security trainings. Note: No data files are included in this entry. This documentation serves as a reference to the raw, unprocessed data files received from CMS.

CCW慢性病症细分数据集概览

慢性病症数据仓库（Chronic Conditions Data Warehouse，以下简称CCW）的慢性病症细分数据集可随核心受益人摘要文件（Master Beneficiary Summary File，以下简称MBSF）基础文件一同申请获取。该细分数据集提供标准化指标，用于判定受益人是否罹患特定慢性病症，其基于联邦医疗保险（Medicare）理赔数据，通过CMS算法生成。慢性病症细分数据集中的每条记录均通过BENE_ID与MBSF基础文件中的受益人信息关联，便于研究人员将慢性病症标识与人口统计学、参保及死亡率数据相结合。

用途与构建方法
CCW慢性病症标识旨在通过应用经过验证的基于理赔数据的算法识别常见慢性疾病，简化研究流程，免去研究人员自行定义编码逻辑的工作。这些算法会在1至3年的指定回溯期内，检索分析住院、门诊、承保商及其他类型的理赔文件，具体回溯时长取决于病症类型。为确保结果可靠，受益人需拥有充足的按服务收费（Fee-for-Service，以下简称FFS）参保记录以覆盖所有相关理赔数据；而参与管理式医疗（Managed Care，以下简称MA/HMO）计划的受益人通常无法获取完整的理赔数据。

可用版本
慢性病症数据仓库（CCW）提供两种标准化慢性病症标识版本：
1. 27项CCW慢性病症标识
该版本由CMS基于经过验证的研究文献与联邦指南开发的算法构建。其通过指定回溯期内的Medicare理赔数据判定慢性病症是否存在，根据年份不同同时使用ICD-9-CM与ICD-10-CM诊断编码。CMS覆盖的数据时段为1999年至2021年。
2. 30项CCW慢性病症标识
2020年，CMS联合专家小组对原有的27项病症算法进行了优化与升级，推出了扩展升级后的30项CCW慢性病症标识，该版本契合最新临床标准，优化了理赔数据处理逻辑，并提升了病症识别准确率。其算法完全基于ICD-10-CM诊断编码（不再使用ICD-9编码）。CMS覆盖的数据时段为2017年至今。

获取与使用规范
由于数据集包含敏感可识别的健康信息，该数据集属于受限使用数据集。因此，本Dataverse条目下无法直接提供数据文件。仅以下人员可申请获取数据：与哈佛大学或其附属机构的首席研究员（Principal Investigator，以下简称PI）合作开展NSAPH相关研究的学生、博士后及研究人员，且需满足以下条件：提交了项目启动表单、已列入获批的机构审查委员会（Institutional Review Board，以下简称IRB）协议名单、并完成了所有要求的数据安全培训。

注意：本条目未包含任何数据文件。本文档仅作为从CMS获取的原始未处理数据文件的参考说明。