Lyons Patrick G, Rojas Juan C, Bewley Alice F, Malone Sara M, Santhosh Lekshmi
Division of Pulmonary, Allergy, and Critical Care Medicine, Department of Medicine, and.
Department of Medical Informatics and Clinical Epidemiology, Oregon Health and Science University, Portland, Oregon.
ATS Sch. 2024 Feb 27;5(2):274-285. doi: 10.34197/ats-scholar.2023-0094OC. eCollection 2024 Jun 1.
Physician communication failures during transfers of patients from the intensive care unit (ICU) to the general ward are common and can lead to adverse events. Efforts to improve written handoffs during these transfers are increasingly prominent, but no instruments have been developed to assess the quality of physician ICU-ward transfer notes.
To collect validity evidence for the modified nine-item Physician Documentation Quality Instrument (mPDQI-9) for assessing ICU-ward transfer note usefulness across several hospitals.
Twenty-four physician raters independently used the mPDQI-9 to grade 12 notes collected from three academic hospitals. , we excluded the "up-to-date" and "accurate" domains, because these could not be assessed without giving the rater access to the complete patient chart. Assessments therefore used the domains "thorough," "useful," "organized," "comprehensible," "succinct," "synthesized," and "consistent." Raters scored each domain on a Likert scale ranging from 1 () to 5 (). The total mPDQI-9 was the sum of these domain scores. The primary outcome was the raters' perceived clinical utility of the notes, and the primary measures of interest were criterion validity (Spearman's ρ) and interrater reliability (intraclass correlation [ICC]).
Mean mPDQI-9 scores by note ranged from 19 (SD = 5.5) to 30 (SD = 4.2). Mean note ratings did not systematically differ by rater expertise (for interaction, = 0.15). The proportion of raters perceiving each note as independently sufficient for patient care (the primary outcome) ranged from 33% to 100% across the set of notes. We found a moderately positive correlation between mPDQI-9 ratings and raters' overall assessments of each note's clinical utility (ρ = 0.48, < 0.001). Interrater reliability was strong; the overall ICC was 0.89 (95% confidence interval [CI], 0.80-0.85), and ICCs were similar among reviewer groups. Finally, Cronbach's α was 0.87 (95% CI, 0.84-0.89), indicating good internal consistency.
We report moderate validity evidence for the mPDQI-9 to assess the usefulness of ICU-ward transfer notes written by internal medicine residents.
在患者从重症监护病房（ICU）转至普通病房的过程中，医生沟通失误很常见，且可能导致不良事件。改善这些转科过程中书面交接班记录的努力日益突出，但尚未开发出评估医生ICU-病房转科记录质量的工具。
收集改良的九项医生记录质量工具（mPDQI-9）的效度证据，以评估其在多家医院评估ICU-病房转科记录有用性方面的作用。
24名医生评估者独立使用mPDQI-9对从三家学术医院收集的12份记录进行评分。我们排除了“最新”和“准确”领域，因为在不给评估者提供完整患者病历的情况下无法对这些领域进行评估。因此，评估使用“全面”“有用”“有条理”“易懂”“简洁”“综合”和“一致”等领域。评估者对每个领域按照从1（差）到5（优）的李克特量表进行评分。mPDQI-9总分是这些领域分数的总和。主要结局是评估者对记录的临床实用性的感知，主要关注的指标是效标效度（斯皮尔曼ρ系数）和评估者间信度（组内相关系数[ICC]）。
每份记录的mPDQI-9平均得分在19（标准差=5.5）至30（标准差=4.2）之间。记录的平均评分在不同评估者专业水平上没有系统性差异（交互作用，P=0.15）。在所有记录中，认为每份记录对患者护理足够独立有用的评估者比例在33%至100%之间。我们发现mPDQI-9评分与评估者对每份记录临床实用性的总体评估之间存在中度正相关（ρ=0.48，P<0.001）。评估者间信度很强；总体ICC为0.89（95%置信区间[CI]，0.80-0.85），各评审组间的ICC相似。最后，克朗巴哈α系数为0.87（95%CI，0.84-0.89），表明内部一致性良好。
我们报告了mPDQI-9在评估内科住院医生书写的ICU-病房转科记录有用性方面有中度效度证据。
