Department of Internal Medicine, Division of Hematology, Oncology & Palliative Care, Virginia Commonwealth University School of Medicine (J.B.C.), Richmond, Virginia, USA.
J Pain Symptom Manage. 2024 Oct;68(4):421-428. doi: 10.1016/j.jpainsymman.2024.07.011. Epub 2024 Jul 18.
While much research has been done regarding "palliative care" using retrospective cohort studies of large datasets, the data sources may not be capturing specialty palliative care services. This article aims to clarify what source data are used in such studies, how specialty palliative care services are determined to have been provided or not, and mismatches between the nature of the data and the interpretation of researchers. Major US data sources that are examined include cancer registries such as the National Cancer Database; health systems' internal data; state and nation-level hospital admissions data; and claims data from Medicare and commercial payers. Problematic studies are common. Many used cancer registry data and mischaracterized palliative intent for a given cancer treatment as "palliative care services." Dozens relied on the diagnosis code for "encounter for palliative care" which lacks adequate validity for use in research. Researchers, peer-reviewers, and research consumers are cautioned about these potential pitfalls that lead to meaningless or misleading research papers. Suggestions are made regarding more rigorous methods and trustworthy data sources and additional research that can lead to consensus among researchers on these issues.
虽然已经有很多关于“姑息治疗”的研究，这些研究采用了回顾性队列研究和大型数据集，但这些数据来源可能无法捕捉到专业姑息治疗服务。本文旨在阐明此类研究中使用了哪些源数据，如何确定是否提供了专业姑息治疗服务，以及数据的性质与研究人员的解释之间的不匹配。本文主要研究了美国的几个主要数据来源，包括癌症登记处，如国家癌症数据库；卫生系统的内部数据；州和国家层面的住院数据；以及医疗保险和商业支付方的索赔数据。有问题的研究很常见。许多研究使用癌症登记处的数据，并将特定癌症治疗的姑息治疗意图错误地描述为“姑息治疗服务”。数十项研究依赖于“姑息治疗就诊”的诊断代码，但该代码在研究中缺乏足够的有效性。研究人员、同行评审人员和研究消费者应注意这些潜在的陷阱，这些陷阱会导致毫无意义或误导性的研究论文。本文就更严格的方法和可靠的数据来源以及可以在这些问题上达成研究人员共识的额外研究提出了建议。
