非急诊儿科急诊就诊的保险公司分类中的种族和民族差异
这项模拟研究基于2016年1月1日至2019年12月31日MarketScan Medicaid数据库中0-18岁Medicaid参保儿童的8,471,386次唯一急诊就诊,应用基于诊断的回顾性理赔算法模拟对可能非急诊就诊的报销削减。结果显示47.7%的就诊被算法识别为可能非急诊,导致研究队列的急诊专业报销总体减少37%。黑人儿童(50.3%)和西班牙裔儿童(49.0%)就诊被识别为非急诊的比例高于白人儿童(45.3%;P < .001)。模拟报销削减后,黑人儿童每次就诊的预期报销比白人儿童低6%,西班牙裔儿童低3%。作者认为,保险公司基于此类算法输出进行财务调整,可能造成不同种族和民族群体间的不均衡报销政策。
要点
- 在8,471,386次儿科Medicaid急诊就诊中,47.7%被基于诊断编码的算法归类为可能非急诊并面临报销削减,导致总体急诊专业报销减少37%。
- 黑人儿童50.3%、西班牙裔儿童49.0%的就诊被算法识别为非急诊,高于白人儿童的45.3%(P < .001)。
- 模拟报销削减后,黑人儿童每次就诊预期报销比白人儿童低6%,西班牙裔儿童低3%。
- 作者认为,保险公司基于此类算法输出进行财务调整,可能造成种族和民族群体间不均衡的报销政策。
适用人群与场景
- 0-18岁Medicaid参保儿童和青少年,数据来自MarketScan Medicaid数据库2016年1月1日至2019年12月31日的急诊就诊。
- 排除缺少出生日期、种族和民族、专业理赔数据、计费复杂度现行程序术语编码的就诊,以及导致住院的就诊。
诊断标准与关键阈值
- 使用基于诊断的回顾性理赔算法将急诊就诊分类为可能非急诊。
研究设计与关键结果
- 研究类型为模拟研究,使用回顾性队列,样本为8,471,386次唯一急诊就诊;43.0%的就诊患者年龄为4-12岁;39.6%为黑人,7.7%为西班牙裔,48.7%为白人。
- 47.7%的就诊被算法识别为可能非急诊并面临报销削减,导致研究队列的急诊专业报销总体减少37%。
- 黑人儿童50.3%、西班牙裔儿童49.0%的就诊被算法识别为非急诊,高于白人儿童的45.3%(P < .001)。
- 模拟报销削减后,黑人儿童每次就诊的预期报销比白人儿童低6%,西班牙裔儿童低3%。
证据类型、等级与局限
- 证据类型:模拟研究,基于回顾性队列。
- 局限:原文未报告局限性。
对床旁实践的改变
- 作者认为,保险公司基于此类算法输出进行财务调整,可能造成不同种族和民族群体间不均衡的报销政策。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
IMPORTANCE: Government and commercial health insurers have recently enacted policies to discourage nonemergent emergency department (ED) visits by reducing or denying claims for such visits using retrospective claims algorithms. Low-income Black and Hispanic pediatric patients often experience worse access to primary care services necessary for preventing some ED visits, raising concerns about the uneven impact of these policies.
OBJECTIVE: To estimate potential racial and ethnic disparities in outcomes of Medicaid policies for reducing ED professional reimbursement based on a retrospective diagnosis-based claims algorithm.
DESIGN, SETTING, AND PARTICIPANTS: This simulation study used a retrospective cohort of pediatric ED visits (aged 0-18 years) for Medicaid-insured children and adolescents appearing in the Market Scan Medicaid database between January 1, 2016, and December 31, 2019. Visits missing date of birth, race and ethnicity, professional claims data, and Current Procedural Terminology codes of billing level of complexity were excluded, as were visits that result in admission. Data were analyzed from October 2021 to June 2022.
MAIN OUTCOMES AND MEASURES: Proportion of ED visits algorithmically classified as nonemergent and simulated per-visit professional reimbursement after applying a current reimbursement reduction policy for potentially nonemergent ED visits. Rates were calculated overall and compared by race and ethnicity.
RESULTS: The sample included 8 471 386 unique ED visits (43.0% by patients aged 4-12 years; 39.6% Black, 7.7% Hispanic, and 48.7% White), of which 47.7% were algorithmically identified as potentially nonemergent and subject to reimbursement reduction, resulting in a 37% reduction in ED professional reimbursement across the study cohort. More visits by Black (50.3%) and Hispanic (49.0%) children were algorithmically identified as nonemergent when compared with visits by White children (45.3%; P < .001). Modeling the impact of the reimbursement reductions across the cohort resulted in expected per-visit reimbursement that was 6% lower for visits by Black children and 3% lower for visits by Hispanic children relative to visits by White children.
CONCLUSIONS AND RELEVANCE: In this simulation study of over 8 million unique ED visits, algorithmic approaches for classifying pediatric ED visits that used diagnosis codes identified proportionately more visits by Black and Hispanic children as nonemergent. Insurers applying financial adjustments based on these algorithmic outputs risk creating uneven reimbursement policies across racial and ethnic groups.
参考来源
- Pomerantz A, De Souza HG, Hall M, et al. Racial and Ethnic Differences in Insurer Classification of Nonemergent Pediatric Emergency Department Visits. JAMA Netw Open. 2023. PMID: 37140920. doi:10.1001/jamanetworkopen.2023.11752
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。