新型冠状病毒肺炎临床数据的统计分析：经验教训、常见错误及避免方法简述
Statistical Analysis of Clinical COVID-19 Data: A Concise Overview of Lessons Learned, Common Errors and How to Avoid Them.
作者信息
Wolkewitz Martin, Lambert Jerome, von Cube Maja, Bugiera Lars, Grodd Marlon, Hazard Derek, White Nicole, Barnett Adrian, Kaier Klaus
机构信息
Institute of Medical Biometry and Statistics, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany.
School of Public Health and Social Work, Queensland University of Technology, Brisbane, QLD, Australia.
出版信息
Clin Epidemiol. 2020 Sep 3;12:925-928. doi: 10.2147/CLEP.S256735. eCollection 2020.
Abstract
By definition, in-hospital patient data are restricted to the time between hospital admission and discharge (alive or dead). For hospitalised cases of COVID-19, a number of events during hospitalization are of interest regarding the influence of risk factors on the likelihood of experiencing these events. The same is true for predicting times from hospital admission of COVID-19 patients to intensive care or from start of ventilation (invasive or non-invasive) to extubation. This logical restriction of the data to the period of hospitalisation is associated with a substantial risk that inappropriate methods are used for analysis. Here, we briefly discuss the most common types of bias which can occur when analysing in-hospital COVID-19 data.
摘要
根据定义，住院患者数据仅限于住院期间至出院（无论存活或死亡）这段时间。对于新冠肺炎住院病例，住院期间的一些事件对于了解风险因素对发生这些事件可能性的影响很重要。对于预测新冠肺炎患者从入院到重症监护的时间，或从开始通气（有创或无创）到拔管的时间也是如此。将数据逻辑上限制在住院期间存在一个重大风险，即可能使用了不恰当的分析方法。在此，我们简要讨论在分析新冠肺炎住院数据时可能出现的最常见偏差类型。
相似文献
1
Statistical Analysis of Clinical COVID-19 Data: A Concise Overview of Lessons Learned, Common Errors and How to Avoid Them.新型冠状病毒肺炎临床数据的统计分析：经验教训、常见错误及避免方法简述
Clin Epidemiol. 2020 Sep 3;12:925-928. doi: 10.2147/CLEP.S256735. eCollection 2020.
2
[Volume and health outcomes: evidence from systematic reviews and from evaluation of Italian hospital data].[容量与健康结果：来自系统评价和意大利医院数据评估的证据]
Epidemiol Prev. 2013 Mar-Jun;37(2-3 Suppl 2):1-100.
3
Interventions for smoking cessation in hospitalised patients.住院患者戒烟干预措施。
Cochrane Database Syst Rev. 2007 Jul 18(3):CD001837. doi: 10.1002/14651858.CD001837.pub2.
4
Adverse drug reactions in an elderly hospitalised population: inappropriate prescription is a leading cause.老年住院患者的药物不良反应：不适当的处方是主要原因。
Drugs Aging. 2005;22(9):767-77. doi: 10.2165/00002512-200522090-00005.
5
[Bias in time delay in ICU admission as a mortality risk factor or "lead time bias"].[重症监护病房入院时间延迟偏差作为死亡风险因素或“领先时间偏差”]
Med Intensiva. 2008 Aug-Sep;32(6):272-6. doi: 10.1016/s0210-5691(08)70954-9.
6
Effectiveness of dexmedetomidine versus propofol on extubation times, length of stay and mortality rates in adult cardiac surgery patients: a systematic review and meta-analysis.右美托咪定与丙泊酚对成人心脏手术患者拔管时间、住院时间和死亡率的影响：一项系统评价和荟萃分析。
JBI Database System Rev Implement Rep. 2018 May;16(5):1220-1239. doi: 10.11124/JBISRIR-2017-003488.
7
Understanding 30-day re-admission after hospitalisation of older patients for diabetes: identifying those at greatest risk.了解老年糖尿病患者住院后 30 天内再入院的情况：确定风险最高的人群。
Med J Aust. 2017 Mar 6;206(4):170-175. doi: 10.5694/mja16.00671.
9
10
Frailty and invasive mechanical ventilation: association with outcomes, extubation failure, and tracheostomy.衰弱与有创机械通气：与结局、拔管失败和气管切开的关系。
Intensive Care Med. 2019 Dec;45(12):1742-1752. doi: 10.1007/s00134-019-05795-8. Epub 2019 Oct 8.
引用本文的文献
1
Interpretation of COVID-19 Epidemiological Trends in Mexico Through Wastewater Surveillance Using Simple Machine Learning Algorithms for Rapid Decision-Making.通过使用简单机器学习算法进行废水监测以实现快速决策来解读墨西哥的新冠疫情流行趋势
Viruses. 2025 Jan 15;17(1):109. doi: 10.3390/v17010109.
2
Association between immunosuppressive medications and COVID-19 hospitalisation and death: a retrospective cohort study.免疫抑制药物与新冠病毒感染住院及死亡之间的关联：一项回顾性队列研究。
BMJ Open. 2024 Dec 26;14(12):e087467. doi: 10.1136/bmjopen-2024-087467.
3
Prioritising deteriorating patients using time-to-event analysis: prediction model development and internal-external validation.使用事件发生时间分析对病情恶化患者进行优先级排序：预测模型的开发与内部-外部验证
Crit Care. 2024 Jul 17;28(1):247. doi: 10.1186/s13054-024-05021-y.
4
Lessons learned: avoiding bias via multi-state analysis of patients' trajectories in real-time.经验教训：通过对患者轨迹进行多状态实时分析避免偏差。
Front Med (Lausanne). 2024 Jun 17;11:1390549. doi: 10.3389/fmed.2024.1390549. eCollection 2024.
5
Methodological biases in observational hospital studies of COVID-19 treatment effectiveness: pitfalls and potential.COVID-19治疗效果的观察性医院研究中的方法学偏倚：陷阱与潜力
Front Med (Lausanne). 2024 Mar 21;11:1362192. doi: 10.3389/fmed.2024.1362192. eCollection 2024.
6
Immunosuppressive Medications and COVID-19 Outcomes in Patients with Noninfectious Uveitis in the Era of COVID-19 Vaccinations.新冠疫苗接种时代非感染性葡萄膜炎患者的免疫抑制药物与新冠病毒病结局
Ophthalmol Sci. 2023 Oct 13;4(2):100411. doi: 10.1016/j.xops.2023.100411. eCollection 2024 Mar-Apr.
7
Impact of the Severe acute respiratory syndrome coronavirus 2 pandemic on mortality associated with healthcare-associated infections.严重急性呼吸综合征冠状病毒2大流行对与医疗保健相关感染相关死亡率的影响。
Antimicrob Steward Healthc Epidemiol. 2023 Aug 29;3(1):e142. doi: 10.1017/ash.2023.409. eCollection 2023.
8
Target trial emulation with multi-state model analysis to assess treatment effectiveness using clinical COVID-19 data.基于多状态模型分析的目标试验模拟，利用临床 COVID-19 数据评估治疗效果。
BMC Med Res Methodol. 2023 Sep 2;23(1):197. doi: 10.1186/s12874-023-02001-8.
9
Comparison of clinical outcomes over time of inpatients with healthcare-associated or community-acquired coronavirus disease 2019 (COVID-19): A multicenter, prospective cohort study.比较医疗机构相关性和社区获得性 2019 冠状病毒病（COVID-19）住院患者随时间推移的临床结局：一项多中心、前瞻性队列研究。
Infect Control Hosp Epidemiol. 2024 Jan;45(1):75-81. doi: 10.1017/ice.2023.143. Epub 2023 Aug 2.
10
Target Trial Emulation Using Hospital-Based Observational Data: Demonstration and Application in COVID-19.利用基于医院的观察性数据进行目标试验模拟：在 COVID-19 中的示范与应用
Life (Basel). 2023 Mar 13;13(3):777. doi: 10.3390/life13030777.
本文引用的文献
1
RETRACTED: Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis.撤回：羟氯喹或氯喹联合或不联合大环内酯类药物治疗新型冠状病毒肺炎：一项多国注册分析
Lancet. 2020 May 22. doi: 10.1016/S0140-6736(20)31180-6.
2
Waste in covid-19 research.新冠疫情研究中的浪费现象。
BMJ. 2020 May 12;369:m1847. doi: 10.1136/bmj.m1847.
3
Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in New York State.纽约州 COVID-19 患者住院死亡率与羟氯喹或阿奇霉素治疗的关联。
JAMA. 2020 Jun 23;323(24):2493-2502. doi: 10.1001/jama.2020.8630.
4
Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal.COVID-19 诊断和预后预测模型：系统评价和批判性评估。
BMJ. 2020 Apr 7;369:m1328. doi: 10.1136/bmj.m1328.
5
Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study.中国武汉成人 COVID-19 住院患者的临床病程和死亡危险因素：一项回顾性队列研究。
Lancet. 2020 Mar 28;395(10229):1054-1062. doi: 10.1016/S0140-6736(20)30566-3. Epub 2020 Mar 11.
6
Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease 2019 Pneumonia in Wuhan, China.中国武汉 2019 年冠状病毒病肺炎患者急性呼吸窘迫综合征和死亡的相关危险因素。
JAMA Intern Med. 2020 Jul 1;180(7):934-943. doi: 10.1001/jamainternmed.2020.0994.
7
Infectious disease consultation for candidaemia.念珠菌血症的感染病会诊。
Lancet Infect Dis. 2020 Feb;20(2):164-165. doi: 10.1016/S1473-3099(19)30745-5.
8
Ribavirin and Interferon Therapy for Critically Ill Patients With Middle East Respiratory Syndrome: A Multicenter Observational Study.利巴韦林和干扰素治疗中东呼吸综合征危重症患者：一项多中心观察性研究。
Clin Infect Dis. 2020 Apr 15;70(9):1837-1844. doi: 10.1093/cid/ciz544.
9
The population-attributable fraction for time-dependent exposures using dynamic prediction and landmarking.基于动态预测和时间标记的时变暴露的人群归因分数。
Biom J. 2020 May;62(3):583-597. doi: 10.1002/bimj.201800252. Epub 2019 Jun 19.
10
Simulation shows undesirable results for competing risks analysis with time-dependent covariates for clinical outcomes.模拟显示，对于具有时间依赖性协变量的临床结局竞争风险分析，结果不理想。
BMC Med Res Methodol. 2018 Jul 16;18(1):79. doi: 10.1186/s12874-018-0535-5.
Zotero 插件