创伤救治费用的机构间差异与死亡率无关:一项全国回顾性队列研究
该回顾性队列研究基于2016-2021年全国住院患者样本,纳入约1066万例创伤住院。主要结局为医院间住院费用和死亡率的变异。医院效应解释费用变异的13%、死亡率变异的3%。高费用中心(风险调整费用最高十分位)收治病情更重患者(ISS 5 vs 3),但调整后与住院死亡率无关(aOR 1.02,95%CI 0.98-1.07),仅与更长住院时间相关(β 0.77,95%CI 0.64-0.90)。严重损伤(ISS≥15)亚组结果一致。
要点
- 医院效应仅解释13%的费用变异,而死亡率变异仅3%,提示费用差异主要非由病例复杂度驱动。
- 高费用中心(HCC)风险调整后住院死亡率无差异(aOR 1.02,95%CI 0.98-1.07),严重损伤亚组亦无差异(aOR 0.95,95%CI 0.90-1.00)。
- HCC与更长住院时间独立相关(β 0.77,95%CI 0.64-0.90),可能反映效率差异而非救治强度获益。
- HCC收治更高 acuity 病例(ISS中位数5 vs 3)且更多为都市教学中心(50.2% vs 37.4%)。
- 证据等级为预后/流行病学Level III,提示需系统层面效率改进而非单纯增加花费。
适用人群与场景
- 2016-2021年全国住院患者样本中因创伤损伤住院的患者,估计约10,661,449例次 admission。
- 严重损伤亚组定义为损伤严重度评分(ISS)≥15。
- 原文未提及排除标准或具体急诊/ICU场景限定。
诊断标准与关键阈值
- 高费用中心(HCC)定义为风险调整后费用最高十分位的医院。
- 严重损伤亚组标准:损伤严重度评分(ISS)≥15。
- 原文未提及诊断标准、评分界值或启动/停止阈值。
处置流程与药物剂量
- 原文未提及
红旗征与不能漏的情况
- 原文未提及
研究设计与关键结果
- 研究类型:全国回顾性队列研究,基于2016-2021年国家住院患者样本,估计10,661,449例创伤住院,8.1%在高费用中心(HCC)。
- 主要终点:医院效应(组内相关系数)解释13%的住院费用变异,仅3%的死亡率变异。
- HCC收治更高 acuity 病例(ISS中位数5 [1-16] vs 3 [1-9],|SMD|=0.22)且更多为都市教学中心(50.2% vs 37.4%,p<0.001)。
- 调整后HCC状态与住院死亡率无关(aOR 1.02,95%CI 0.98-1.07),但与更长住院时间相关(β 0.77,95%CI 0.64-0.90)。
- 严重损伤(ISS≥15)亚组:医院效应解释12%费用变异和2%死亡率变异,HCC状态与死亡率仍无关(aOR 0.95,95%CI 0.90-1.00)。
证据类型、等级与局限
- 证据类型:回顾性队列研究(全国住院患者样本数据库分析)。
- 证据等级:原文标注Level of Evidence: Prognostic/Epidemiological; Level III。
- 局限:回顾性设计、基于行政数据库、费用为估算值(成本收费比转换并通胀调整至2021美元)、原文未提及单中心或多中心限制(实为全国样本但基于数据库)。
对床旁实践的改变
- 提示创伤救治费用在医院间差异显著但死亡率变异极小,高费用医院未显示风险调整生存获益,应关注系统效率而非单纯增加花费。
- 高费用中心收治更高 acuity 患者且住院时间更长,临床中不应将高费用等同于高质量救治。
- 适用人群:创伤住院患者;不适用人群原文未提及。
- 常见陷阱:将观察性费用-死亡率关联误解为因果,或忽略病例 mix 差异。
指南或共识出处
- 原文未提及具体指南或共识名称;研究发表于J Trauma Acute Care Surg。
以上提炼依据论文摘要(全文未开放获取,或许可不允许翻译转载);全文请点「阅读原文」。
英文原文(摘要)· 点击展开
BACKGROUND: Trauma care costs have risen nationally, yet whether higher inpatient spending translates into better outcomes remains unclear. We evaluated contemporary interhospital variation in trauma hospitalization costs and mortality and assessed whether higher-cost hospitals achieved lower risk-adjusted mortality.
METHODS: This was a retrospective cohort study of hospitalizations entailing traumatic injury identified in the 2016-2021 National Inpatient Sample. The primary outcomes were interfacility variation in hospitalization costs and mortality. Estimated hospitalization costs were derived using cost-to-charge ratios and inflation-adjusted to 2021 dollars. Hierarchical linear and logistic regression models were used to quantify hospital-attributable variation in costs and mortality via intraclass correlation coefficients. Costs were log-transformed for modeling, and hospitals in the highest decile of risk-adjusted costs were designated High-Cost Centers (HCCs). Multivariable models assessed associations of HCC status with inpatient mortality and length of stay (LOS). A subgroup analysis examined patients with severe injuries [Injury Severity Score (ISS) ≥15].
RESULTS: Among an estimated 10,661,449 admissions, 8.1% were at HCC. Intraclass correlation coefficient analysis demonstrated that hospital effects accounted for 13% of the variation in inpatient costs but only 3% of the variation in mortality. HCC treated a higher-acuity case mix (ISS: 5 [1-16] vs. 3 [1-9], |SMD|=0.22) and were more commonly metropolitan teaching centers (50.2 vs. 37.4%, p<0.001). After adjustment, HCC status was not associated with inpatient mortality (adjusted odds ratio, 1.02; 95% CI, 0.98-1.07) but was associated with longer LOS (β, 0.77; 95% CI, 0.64-0.90). In severe injuries (ISS ≥15), hospital effects explained 12% of cost variation and 2% of mortality variation, and HCC status remained unassociated with mortality (adjusted odds ratio, 0.95; 95% CI, 0.90-1.00).
CONCLUSIONS: Trauma hospitalization costs vary substantially across US hospitals, whereas hospital-level variation in mortality is minimal. High-cost hospitals cared for higher-acuity patients and had longer LOS, yet did not demonstrate improved risk-adjusted survival, highlighting opportunities to improve value through system-level efficiency efforts.
LEVEL OF EVIDENCE: Prognostic/Epidemiological; Level III.
参考来源
- Coaston T, Mallick S, Gao Z, et al. Interfacility variation in trauma care costs without differences in mortality: A national retrospective cohort study. J Trauma Acute Care Surg. 2026. PMID: 42479808. doi:10.1097/TA.0000000000005129
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。