Klompas Michael, Yokoe Deborah S
Infection Control Department, Channing Laboratory, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA.
Clin Infect Dis. 2009 May 1;48(9):1268-75. doi: 10.1086/597591.
Health care providers, quality advocates, consumers, and legislators are increasingly focused on the prevention of health care-associated infections. Accurate surveillance is essential to identify areas for improvement and to measure the impact of infection prevention initiatives. Conventional surveillance definitions, however, are complicated, costly to apply, and prone to both intentional and unintentional misclassification. Algorithmic analysis of electronic health data is a promising alternative to conventional surveillance. Algorithms that seek combinations of diagnosis codes, microbiological analysis results, and/or antimicrobial dispensing can identify health care-associated infections with sensitivities and positive predictive values that often match or surpass those of conventional surveillance. The efficiency and objectivity of these methods make them promising candidates for more manageable and meaningful benchmarking within and between facilities.
医疗服务提供者、质量倡导者、消费者和立法者越来越关注医疗保健相关感染的预防。准确的监测对于确定改进领域和衡量感染预防举措的影响至关重要。然而，传统的监测定义复杂、应用成本高，并且容易出现有意和无意的错误分类。对电子健康数据进行算法分析是传统监测的一种有前途的替代方法。通过寻求诊断代码、微生物分析结果和/或抗菌药物配发组合的算法，可以识别医疗保健相关感染，其敏感性和阳性预测值通常与传统监测相当或更高。这些方法的效率和客观性使其成为设施内部和之间更易于管理和更有意义的基准测试的有前途的候选方法。
