A novel approach to measuring costs and efficiency: Lung nodules as a case study

每年有150多万退伍军人接受胸部X光或CT扫描，以帮助评估常见症状 比如呼吸急促。成像质量的提高导致了近一半的常规扫描 发现可能引发肺癌的偶发小结节。尽管临床指南 对于如何管理这些不确定的结节，我们不知道提供者是否遵循建议 如果这些患者接受了太少或太多的后续成像和测试。 在这个项目中，我们建议确定2011财年的所有退伍军人(估计有200,000名患者) 确诊为肺结节，并确定他们是否接受了指南建议的治疗 随后24个月。先前的一项胸部成像研究报告称，一个可疑的发现会产生超过1300美元的收入。 在年度后续费用中。这将转化为每年超过5亿美元的重复成像和测试 VHA因偶然发现而产生的费用。由于这一成本可能是巨大的，我们将对两者进行衡量 患者是否接受了指南建议的护理，以及所有肺结节护理的费用 他或她在这24个月期间收到的活动。初步证据表明，可能存在广泛的 当发现偶然的肺结节时，退伍军人所接受的护理的变化。这个项目的目标是 评估退伍军人接受的护理质量，并确定似乎提供最佳质量的设施 以最低的成本提供护理。我们的长期目标是确定在这些设施和 将这些做法传播到整个VHA。 为了识别患有偶发肺结节的退伍军人并测量其依从性 根据指南，我们将访问放射学文本报告，这些报告现已在VHA全国范围内提供。这项工作将 使用自然语言处理来提取有关特征的详细信息 从正文中报告每一个肺结节，以及作为患者报告的任何结节特征的变化 在24个月期间进行重复成像。 我们提出两个研究目标： 目标1：使用自然语言处理将遵守Fleishner指南的措施付诸实施 管理肺结节，并评估VHA指南遵从性的差异。 目标2：衡量管理肺结节的使用率和费用，并评估使用率和费用的差异 在威斯康星州。

Over 1.5 million Veterans receive a chest x-ray or CT scan each year to help evaluate common symptoms such as shortness of breath. Improvements in imaging quality have led to nearly half of these routine scans finding small incidental lung nodules that are potentially worrisome for lung cancer. Although clinical guidelines exist for how to manage these indeterminate nodules, we do not know if providers follow recommended guidelines, or if these patients receive too little or too much follow-up imaging and testing. In this project we propose to identify all Veterans in FY2011 (estimated to be 200,0000 patients) who were identified with a lung nodule and determine whether they received guideline recommended care over the subsequent 24 months. A prior study of chest imaging reported that a suspicious finding generates over $1300 in annual follow-up costs. This would translate to over $500 million dollars a year in repeat imaging and testing costs for VHA due to incidental findings. Because this cost is potentially enormous, we will measure both whether a patient received guideline recommended care, as well as the cost of all the lung nodule care activities he or she received over this 24 month period. Preliminary evidence suggests there may be wide variation in the care Veterans receive when an incidental lung nodule is found. The goal of this project is to assess the quality of care Veterans receive and identify facilities that appear to be delivering the best quality care at the lowest cost. Our long term objective is to identify the approaches used at those facilities and disseminate those practices across VHA. In order to identify Veterans with an incidental lung nodule and measure adherence with recommended guidelines, we will access radiology text reports, which are now available nationally across VHA. This work will be accomplished using natural language processing to extract detailed information about the characteristics of each lung nodule from the text reports, as well as any changes reported in nodule characteristics as patients undergo repeated imaging during the 24 month period. We propose two study aims: Aim 1: Use natural language processing to operationalize a measure of adherence to Fleishner guidelines for managing lung nodules, and assess variation in guideline adherence across VHA. Aim 2: Measure utilization and costs of managing lung nodules, and assess variation in utilization and costs across VHA.