Keely Erin, Myers Kathryn, Dojeiji Suzan, Campbell Craig
Department of Medicine, University of Ottawa, Canada.
BMC Med Educ. 2007 May 22;7:13. doi: 10.1186/1472-6920-7-13.
Written correspondence is one of the most important forms of communication between health care providers, yet there is little feedback provided to specialists. The objective of this study was to determine the feasibility and satisfaction of a peer assessment program on consultation letters and to determine inter-rater reliability between family physicians and specialists.
A rating scale of nine 5-point Likert scale items including specific content, style items, education value of the letter and an overall rating was developed from a previous validated tool. Nine Internal Medicine specialists/subspecialists from two tertiary care centres submitted 10 letters with patient and physician identifiers removed. Two Internal Medicine specialists, and 2 family physicians from the other centre rated each letter (to protect writer anonymity). A satisfaction survey was sent to each writer and rater after collation of the results. A follow-up survey was sent 6-8 months later.
There was a high degree of satisfaction with the process and feedback. The rating scale information was felt to be useful and appropriate for evaluating the quality of consultation letters by 6/7 writers. 5/7 seven writers felt that the feedback they received resulted in immediate changes to their letters. Six months later, 6/9 writers indicated they had maintained changes in their letters. Raters rank ordered letters similarly (Cronbach's alpha 0.57-0.84) but mean scores were highly variant. At site 1 there were significant differences in scoring brevity (p < 0.01) between family physician and specialist raters; whereas, at site 2 there were differences in scoring of history (p < 0.01), physical examination (p < 0.01) and educational value (p < 0.01) of the letter.
Most participants found peer assessment of letters feasible and beneficial and longstanding changes occurred in some individuals. Family physicians and specialists appear to have different expectations on some items. Further studies on reliability and validity, with a larger sample, are required before high stakes professional assessments include consultation letters.
书面通信是医疗服务提供者之间最重要的沟通形式之一，但针对专科医生的反馈却很少。本研究的目的是确定同行评估计划对会诊信的可行性和满意度，并确定家庭医生和专科医生之间的评分者间信度。
从先前经过验证的工具中开发了一个包含九个5点李克特量表项目的评分量表，包括具体内容、风格项目、信件的教育价值和总体评分。来自两个三级医疗中心的九名内科专科医生/亚专科医生提交了10封去除了患者和医生标识符的信件。来自另一个中心的两名内科专科医生和两名家庭医生对每封信进行评分（以保护作者的匿名性）。在整理结果后，向每位作者和评分者发送了一份满意度调查问卷。6至8个月后发送了一份随访调查问卷。
对该过程和反馈有高度满意度。6/7的作者认为评分量表信息对于评估会诊信的质量是有用且合适的。5/7的作者认为他们收到的反馈使他们立即对信件进行了修改。六个月后，6/9的作者表示他们保持了信件中的修改。评分者对信件的排名相似（克朗巴哈系数为0.57 - 0.84），但平均得分差异很大。在地点1，家庭医生和专科医生评分者在简洁性评分上存在显著差异（p < 0.01）；而在地点2，信件的病史评分（p < 0.01）、体格检查评分（p < 0.01）和教育价值评分（p < 0.01）存在差异。
大多数参与者认为对信件进行同行评估是可行且有益的，并且一些人发生了长期变化。家庭医生和专科医生在某些项目上似乎有不同的期望。在高风险专业评估纳入会诊信之前，需要对更大样本进行信度和效度的进一步研究。
