中国全科医学 ›› 2024, Vol. 27 ›› Issue (04): 460-467.DOI: 10.12114/j.issn.1007-9572.2023.0005

• 论著·慢性病共病研究 • 上一篇    下一篇

基于行政管理数据识别慢性病发病病例背景下确定洗脱期时长最佳策略的系统综述

杨文怡, 王敬鑫, 艾丽梅, 万霞*()   

  1. 100005 北京市,中国医学科学院基础医学研究所 北京协和医学院基础学院
  • 收稿日期:2023-01-05 修回日期:2023-04-18 出版日期:2024-02-05 发布日期:2023-11-09
  • 通讯作者: 万霞

  • 作者贡献:杨文怡提出研究选题方向,负责文献检索和整理,并撰写论文初稿;王敬鑫进行文献数据提取和信息整理;艾丽梅负责论文的修订;万霞负责文章的质量控制及审校,对文章整体负责;所有作者确认了论文的最终稿。
  • 基金资助:
    美国中华医学基金会项目(15-208)

Optimal Strategies for Determining the Duration of Washout Period in the Context of Identifying Chronic Disease Onset Cases Based on Administrative Data: a Systematic Review

YANG Wenyi, WANG Jingxin, AI Limei, WAN Xia*()   

  1. Institute of Basic Medical Sciences, Chinese Academy of Medical Science & School of Basic Medicine, Peking Union Medical College, Beijing 100005, China
  • Received:2023-01-05 Revised:2023-04-18 Published:2024-02-05 Online:2023-11-09
  • Contact: WAN Xia

摘要: 背景 使用行政管理数据时,确立清晰、适当的慢性病洗脱期时长是正确确定反复就医的慢性病患者发病时点、确定新发病例的基础。 目的 通过系统文献回顾,综述确定洗脱期时长的方法,以期为我国研究者后续使用行政管理数据识别慢性病新发病例时确认洗脱期长短、正确识别新发病例提供思路。 方法 于2021年10月,系统检索PubMed、Web of Science、EmBase、中国知网、维普中文科技期刊全文数据库、万方知识服务平台,获取有关利用行政管理数据探究慢性病发病、患病情况的文献,检索时限均为建库至2022-10-01。由两名研究者独立筛选文献并提取相关信息,并采用定性研究报告评价标准(SRQR)评价文献方法学质量后,使用描述性分析法总结洗脱期时长的确定方法。 结果 共纳入26篇文献,纳入文献的SRQR评分均≥15分,方法学质量较好。文献所使用的数据主要来自加拿大、美国、澳大利亚等行政管理数据完整、丰富的国家(地区),聚焦的疾病包括糖尿病、肿瘤、精神分裂症等多种慢性病。研究指出,设定合适的洗脱期时长是准确识别发病病例的基础。目前,文献中确定洗脱期时长的方法主要包括直接限定法、一致性检验法和逆向生存函数法三大类,其中最常用的方法是直接限定法,逆向生存曲线法的使用率相对较低。 结论 直接限定法、一致性检验法和逆向生存函数法均有相应的优势和局限性,方法的选择标准、判断标准和稳定性有待进一步探究。

关键词: 慢性病, 行政管理数据, 医疗保险数据, 洗脱期, 发病率, 患病率

Abstract:

Background

When using administrative data, the establishment of clear and appropriate duration of chronic disease washout period is the basis for correctly identifying the point of onset of chronic disease in patients with recurrent visits and identifying new cases.

Objective

To review the methods for determining the duration of washout period by a systematic review of literature, in order to provide ideas for Chinese researchers to confirm the duration of washout period and correctly identify new cases with subsequent use of administrative data to identify new cases of chronic diseases.

Methods

In October 2021, PubMed, Web of Science, EmBase, CNKI, CQVIP and Wanfang Knowledge Service Platform were systematically searched for literature on the use of administrative data to explore the incidence and prevalence of chronic diseases from inception to 2022-10-01. Two researchers independently screened literatures and extracted relevant information, evaluated the quality of literature by using the Standards for Reporting Qualitative Research (SRQR) scale, and summarized the determination method of the duration of washout period.

Results

A total of 26 papers were included, and the SRQR scores of the included literature were all≥15, indicating good methodological quality. The data used in the literature were mainly from Canada, the United States, Australia and other countries (regions) with complete and abundant administrative data, and focused on a variety of chronic diseases, including diabetes, tumor, schizophrenia and other chronic diseases. The study pointed out that setting an appropriate washout period duration is the basis for accurate identification of onset cases. The methods used to determine the duration of washout period in literature were mainly divided into three major categories, including direct restriction method, consistency test method and retrograde survival function method. Among them, the most commonly used method was direct restriction method, while the retrograde survival function method had a relatively low usage rate.

Conclusion

Direct restriction method, consistency test method and retrograde survival function method all have corresponding advantages and limitations. The selection, judgment criteria and stability of the method need to be further explored.

Key words: Chronic disease, Administrative data, Insurance data, Washout period, Incidence, Prevalence