Multidisciplinary Epidemiology and Translational Research in Intensive Care, Division of Pulmonary and Critical Care Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA.
Respir Med. 2011 Oct;105(10):1501-6. doi: 10.1016/j.rmed.2011.04.012. Epub 2011 May 14.
Participant recruitment is an important process in successful conduct of randomized controlled trials. To facilitate enrollment into a National Institutes of Health-sponsored clinical trial involving patients with chronic obstructive pulmonary disease (COPD), we developed and prospectively validated an automated electronic screening tool based on boolean free-text search of admission notes in electronic medical records. During a 2-week validation period, all patients admitted to prespecified general medical services were screened for eligibility by both the electronic screening tool and a COPD nurse. Group discussion was the gold standard for confirmation of true-positive results. Compared with the gold standard, electronic screening yielded 100% sensitivity, 92% specificity, 100% negative predictive value, and 72% positive predictive value. Compared with traditional manual screening, electronic screening demonstrated time-saving potential of 76%. Thus, the electronic screening tool accurately identifies potential study subjects and improves efficiency of patient accrual for a clinical trial on COPD. This method may be expanded into other institutional and clinical settings.
参与者招募是成功进行随机对照试验的重要过程。为了促进参与一项由美国国立卫生研究院(NIH)赞助的涉及慢性阻塞性肺疾病(COPD)患者的临床试验,我们开发并前瞻性验证了一种基于电子病历中入院记录的布尔自由文本搜索的自动化电子筛选工具。在为期两周的验证期内,所有被收入指定的普通医疗服务的患者均通过电子筛选工具和 COPD 护士进行了资格筛选。小组讨论是确认真阳性结果的金标准。与金标准相比,电子筛选的敏感性为 100%,特异性为 92%,阴性预测值为 100%,阳性预测值为 72%。与传统的手动筛选相比,电子筛选具有节省 76%时间的潜力。因此,电子筛选工具可以准确识别潜在的研究对象,提高 COPD 临床试验的患者入组效率。这种方法可以扩展到其他机构和临床环境。