文献速递创伤与烧伤

问卷与访谈评估创伤性脑损伤后整体功能结局的比较

Questionnaires vs Interviews for the Assessment of Global Functional Outcomes After Traumatic Brain Injury.
PubMed 文献 · JAMA Netw Open · 2021-11-01
一句话结论在创伤性脑损伤患者中,问卷与访谈两种GOSE评估方式在3个月和6个月时一致性良好(加权κ分别为0.77和0.82),支持在具有实际优势的研究中使用问卷。

该队列研究基于CENTER-TBI项目2014年12月至2017年12月纳入的≥16岁、有CT检查指征的TBI患者,在伤后3个月和6个月同时完成访谈和问卷两种GOSE评估。3个月时994例(26.9%)完成两种评估,6个月时628例(17.0%)完成。两种方法在3个月(加权κ=0.77;95%CI 0.73-0.80)和6个月(加权κ=0.82;95%CI 0.78-0.86)均一致性良好。访谈比问卷记录更多工作问题、更少社交休闲限制和更多症状,但两种方法与基线因素及患者报告结局的相关强度相似。

要点

  • 在伤后3个月和6个月,访谈与问卷两种GOSE评估方法的一致性良好,加权κ分别为0.77(95%CI 0.73-0.80)和0.82(95%CI 0.78-0.86)。
  • 条目水平上,日常活动独立性部分的一致性良好(两个时间点κ=0.70-0.79),而关系、症状等主观功能部分的一致性中等(两个时间点κ=0.41-0.51)。
  • 与问卷相比,访谈在3个月记录更多工作问题(294[30.5%] vs 233[24.2%])、更少社交和休闲活动限制(330[33.8%] vs 431[44.1%])、更多症状(524[53.6%] vs 324[33.1%]);
  • 访谈者有时基于判断而非访谈评分规则给出总体评级,尤其对于可能预后不良的TBI患者;但两种评估方式与基线因素(ρ=-0.13至0.42)及患者报告结局(ρ=0.29至0.65)的相关强度相似。

适用人群与场景

  • 年龄≥16岁、患有创伤性脑损伤且有计算机断层扫描(CT)成像临床指征的患者,纳入CENTER-TBI项目(2014年12月至2017年12月)。
  • 在伤后3个月和6个月随访时同时完成访谈和问卷两种GOSE评估的个体。

研究设计与关键结果

  • 研究类型:基于CENTER-TBI项目的队列研究;纳入3691例合格个体,其中994例(26.9%)在3个月完成两种GOSE评估(654例[65.8%]男性;中位年龄53岁[IQR 33-66岁]),628例(17.0%)在6个月完成(409例[65.1%]男性;中位年龄51岁[IQR 31-64岁])。
  • 主要终点:访谈与问卷GOSE评分在3个月(加权κ=0.77;95%CI 0.73-0.80)和6个月(加权κ=0.82;95%CI 0.78-0.86)均一致性良好。
  • 条目水平一致性:日常活动独立性部分在两个时间点κ=0.70-0.79(良好);关系、症状等主观功能部分在两个时间点κ=0.41-0.51(中等)。
  • 访谈 vs 问卷:3个月工作问题294例(30.5%) vs 233例(24.2%),社交和休闲活动限制330例(33.8%) vs 431例(44.1%),症状524例(53.6%) vs 324例(33.1%);6个月工作问题161例(26.8%) vs 136例(22.7%),社交和休闲活动限制179例(29.7%) vs 219例(36.4%),症状291例(48.4%) vs 179例(29.8%)。
  • 两种评估方式与基线因素的相关强度(ρ=-0.13至0.42)及与患者报告结局的相关强度(ρ=0.29至0.65)相似。

证据类型、等级与局限

  • 证据类型:队列研究。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为,研究结果支持在问卷形式相比访谈具有显著实际优势的研究中使用问卷。
  • 访谈者有时基于判断而非访谈评分规则给出总体评级,尤其对于可能预后不良的TBI患者,提示访谈评估可能存在评分规则执行不一致的情况。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「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: An interview is considered the gold standard method of assessing global functional outcomes in clinical trials among patients with acute traumatic brain injury (TBI). However, several multicenter clinical trials have used questionnaires completed by a patient or caregiver to assess the primary end point.

OBJECTIVE: To examine agreement between interview and questionnaire formats for assessing TBI outcomes and to consider whether an interview has advantages.

DESIGN, SETTING, AND PARTICIPANTS: This cohort study used data from patients enrolled in the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) project from December 2014 to December 2017. Data were analyzed from December 2020 to April 2021. Included patients were aged 16 years or older with TBI and a clinical indication for computed tomography imaging. Outcome assessments were completed using both an interview and a questionnaire at follow-up 3 and 6 months after injury.

EXPOSURES: Traumatic brain injury of all severities.

MAIN OUTCOMES AND MEASURES: Ratings on the Glasgow Outcome Scale-Extended (GOSE) administered as a structured interview rated by an investigator and as a questionnaire completed by patients or caregivers and scored centrally were compared, and the strength of agreement was evaluated using weighted κ statistics. Secondary outcomes included comparison of different sections of the GOSE assessments and the association of GOSE ratings with baseline factors and patient-reported mental health, health-related quality of life, and TBI symptoms.

RESULTS: Among the 3691 eligible individuals in the CENTER-TBI study, both GOSE assessment formats (interview and questionnaire) were completed by 994 individuals (26.9%) at 3 months after TBI (654 [65.8%] male; median age, 53 years [IQR, 33-66 years]) and 628 (17.0%) at 6 months (409 [65.1%] male; median age, 51 years [IQR, 31-64 years]). Outcomes of the 2 assessment methods agreed well at both 3 months (weighted κ, 0.77; 95% CI, 0.73-0.80) and 6 months (weighted κ, 0.82; 95% CI, 0.78-0.86). Furthermore, item-level agreement between the 2 methods was good for sections regarding independence in everyday activities (κ, 0.70-0.79 across both time points) and moderate for sections regarding subjective aspects of functioning such as relationships and symptoms (κ, 0.41-0.51 across both time points). Compared with questionnaires, interviews recorded more problems with work (294 [30.5%] vs 233 [24.2%] at 3 months and 161 [26.8%] vs 136 [22.7%] at 6 months), fewer limitations in social and leisure activities (330 [33.8%] vs 431 [44.1%] at 3 months and 179 [29.7%] vs 219 [36.4%] at 6 months), and more symptoms (524 [53.6%] vs 324 [33.1%] at 3 months and 291 [48.4%] vs 179 [29.8%] at 6 months). Interviewers sometimes assigned an overall rating based on judgment rather than interview scoring rules, particularly for patients with potentially unfavorable TBI outcomes. However, for both formats, correlations with baseline factors (ρ, -0.13 to 0.42) and patient-reported outcomes (ρ, 0.29 to 0.65) were similar in strength.

CONCLUSIONS AND RELEVANCE: In this cohort study, GOSE ratings obtained by questionnaire and interview methods were in good agreement. The similarity of associations of the ratings obtained by both GOSE methods with baseline factors and other TBI outcome measures suggests that despite some apparent differences, the core information collected by both interviews and questionnaires was similar. The findings support the use of questionnaires in studies in which this form of contact may offer substantial practical advantages compared with interviews.

来源:JAMA Netw Open,查看原文。版权归原作者所有。

参考来源

  1. Horton L, Rhodes J, Menon DK, et al. Questionnaires vs Interviews for the Assessment of Global Functional Outcomes After Traumatic Brain Injury. JAMA Netw Open. 2021. PMID: 34762111. doi:10.1001/jamanetworkopen.2021.34121

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。