Validation of the Functional Comorbidity Index for Use with Workers' Compensation Data: Best Practices for Predicting Work-Related and Functional Outcomes

项目总结/摘要 项目摘要/摘要 大约一半的美国工人报告至少有一种慢性疾病;大约四分之一的人报告多次发病。 慢性病工人的卫生保健利用率较高，健康和就业状况较差 成果。慢性病与许多职业健康研究项目有关，无论是作为主要的 研究重点，或调整并发症负担;然而，当依赖于 行政工人的赔偿（WC）数据。Functional Comorbidity Index（FCI）是一种工具， 特别适合于工作人群，尚未被验证用于专门与WC数据。 关于共同调整的重要性和方法， 当研究工人的结果，部分原因是不确定的程度，慢性疾病是 在WC数据库中捕获，这些数据库通常用于大规模研究工人的结果。因此，有一个 迫切需要一种经验证可用于WC数据的慢性疾病/合并症工具。整体 本提案的目的是利用WC数据，评估并最大化基于WC的FCI的预测有效性 与华盛顿州和俄亥俄州受伤工人的大型前瞻性纵向调查有关。目标1： 评估使用基于WC的诊断代码的有效性：（1A）识别个体慢性疾病，以及 (1B)构建基于WC的FCI。对于18种FCI疾病中的每一种， 将通过WC数据识别与自我报告进行比较。基于WC的FCI将与 自我报告的FCI，评估：（1）同时效度（一致性），（2）预测效度的工作相关， 功能结果，和（3）控制混淆的效用。目标2：评估优化准确性的方法 慢性病识别和FCI预测有效性时，使用WC数据。比较将包括 改变（1）WC数据源，（2）测量时间范围，和（3）FCI构建方法。这 研究是创新的，因为它将提供迄今为止无法获得的数据（1）的可行性和敏感性， 通过WC数据识别慢性疾病，以及（2）使用 带有WC数据的FCI仪器。预期输出包括提供经验证的基于WC的FCI-可能 加权或以其他方式修改，以公共领域，沿着的最佳方法的描述 实践和局限性。这一贡献将是重大的，因为它将有助于实现 改善慢性病监测和研究，以及更加注重在 基于WC的研究，并对局限性进行适当的解释和报告。反过来，预期的增长 在大规模的慢性病研究将支持长期目标的质量改善，在厕所相关的 卫生保健、更健康的劳动力、更高的生产率、更低的WC成本和更好的就业结果。

Project Summary/Abstract oject Summary/Abstract Roughly half of U.S. workers report at least one chronic condition; roughly a quarter report multimorbidity. Workers with chronic conditions have higher health care utilization, and poorer health and employment outcomes. Chronic conditions are relevant to many occupational health research projects, either as the primary study focus, or to adjust for comorbidity burden; however, they are challenging to measure when relying on administrative workers’ compensation (WC) data. The Functional Comorbidity Index (FCI), an instrument particularly well-suited to working populations, has not yet been validated for use specifically with WC data. There are significant knowledge gaps regarding the importance of—and methods for—comorbidity adjustment when studying worker outcomes, due in part to uncertainty about the extent to which chronic conditions are captured in the WC databases that are often used to study worker outcomes at scale. Consequently, there is a pressing need for a chronic condition/comorbidity instrument validated for use with WC data. The overall objective of this proposal is to assess and maximize predictive validity of a WC-based FCI, using WC data linked to a large prospective longitudinal survey of injured workers in Washington State and Ohio. Aim 1: Assess the validity of using WC-based diagnosis codes for: (1A) identifying individual chronic conditions, and (1B) constructing a WC-based FCI. For each of the 18 individual FCI conditions, prevalence and concordance will be calculated, comparing identification via WC data to self-report. The WC-based FCI will be compared to the self-report FCI, assessing: (1) concurrent validity (concordance), (2) predictive validity for work-related and functional outcomes, and (3) utility for control of confounding. Aim 2: Assess methods of optimizing accurate chronic condition identification and FCI predictive validity when using WC data. Comparisons will include varying (1) WC data sources, (2) measurement timeframes, and (3) approaches to FCI construction. This research is innovative because it will provide heretofore unavailable data on (1) the feasibility and sensitivity of identifying chronic conditions via WC data, and (2) the performance, validity, and best practices for use of the FCI instrument with WC data. Expected outputs include contributing a validated WC-based FCI—potentially weighted or otherwise modified—to the public domain, along with a description of best methodological practices and limitations. This contribution will be significant because it will enable the expected outcomes of improved chronic condition surveillance and research, as well as increased focus on comorbidity adjustment in WC-based research, with appropriate interpretation and reporting of limitations. In turn, the expected increase in large-scale chronic condition research will support longer-term goals of quality improvement in WC-related health care, a healthier workforce, higher productivity, lower WC costs, and improved employment outcomes.