Benin Andrea L, Vitkauskas Grace, Thornquist Elizabeth, Shapiro Eugene D, Concato John, Aslan Mihaela, Krumholz Harlan M
Robert Wood Johnson Clinical Scholars Program, Yale University School of Medicine, New Haven, Connecticut, USA.
Med Care. 2005 Jul;43(7):691-8. doi: 10.1097/01.mlr.0000167185.26058.8e.
We sought to evaluate the validity of retrieving data from a commercial, outpatient electronic medical record (EMR) to assess the management of pharyngitis, an important measure of quality of healthcare in pediatrics and a new measure for the Health Plan Employer Data and Information Set (HEDIS).
For children ages 3-18 years, we electronically identified clinical encounters with diagnoses of pharyngitis using 3 different strategies (an EMR-based strategy, an administrative data-based strategy, and a reference strategy which used medical record review). By each strategy, we calculated the proportion of episodes of pharyngitis during 1 year for which management of pharyngitis adhered to published guidelines.
Among 479 total episodes of pharyngitis, 434 (91%) were from the EMR-based strategy and 281 (59%) from the administrative data-based strategy. Review of the records (the reference strategy) found that 391 of 479 (82%) were confirmed episodes of pharyngitis. A diagnostic test for group A streptococcus (GAS) was performed at 310 of 434 (71%) of episodes identified by the EMR-based strategy and at 214 of 281 (76%) of episodes by the administrative data-based strategy (P = 0.045). By the reference strategy, a diagnostic test was done in 301 of 391 (77%); more than at episodes found by the EMR-based strategy (71%, P < 0.001).
The EMR-based strategy resulted in a statistically different proportion of episodes having diagnostic testing for GAS compared with a reference strategy. Complete evaluations to validate strategies for extracting data from electronic databases are necessary before assuming that measures of quality of care will be the same regardless of the source of data.
我们试图评估从商业门诊电子病历（EMR）中检索数据以评估咽炎管理的有效性，咽炎管理是儿科医疗质量的一项重要指标，也是健康计划雇主数据与信息集（HEDIS）的一项新指标。
对于3至18岁的儿童，我们使用3种不同策略（基于EMR的策略、基于行政数据的策略以及使用病历审查的参考策略）以电子方式识别诊断为咽炎的临床就诊情况。通过每种策略，我们计算了1年中咽炎发作且其管理符合已发表指南的比例。
在总共479例咽炎发作中，434例（91%）来自基于EMR的策略，281例（59%）来自基于行政数据的策略。病历审查（参考策略）发现，479例中有391例（82%）为确诊的咽炎发作。在基于EMR的策略识别出的434例发作中的310例（71%）以及基于行政数据的策略识别出的281例发作中的214例（76%）进行了A组链球菌（GAS）诊断检测（P = 0.045）。按照参考策略，在391例中的301例（77%）进行了诊断检测；比基于EMR的策略发现的发作（71%）更多（P < 0.001）。
与参考策略相比，基于EMR的策略导致进行GAS诊断检测的发作比例在统计学上存在差异。在假定无论数据来源如何护理质量指标都相同之前，有必要进行全面评估以验证从电子数据库提取数据的策略。
