Cooper B S, Medley G F, Scott G M
Department of Biological Sciences, University of Warwick, Coventry, UK.
J Hosp Infect. 1999 Oct;43(2):131-47. doi: 10.1053/jhin.1998.0647.
A simple mathematical model is developed for the spread of hand-borne nosocomial pathogens such as Staphylococcus aureus within a general medical-surgical ward. In contrast to previous models a stochastic approach is used. Computer simulations are used to explore the properties of the model, and the results are presented in terms of the pathogen's successful introduction rate, ward-level prevalence, and colonized patient-days, emphasizing the general effects of changes in management of patients and carers. Small changes in the transmissibility of the organism resulted in large changes in all three measures. Even small increases in the frequency of effective handwashes were enough to bring endemic organisms under control. Reducing the number of colonized patients admitted to the ward was also an effective control measure across a wide range of different situations. Increasing surveillance activities had little effect on the successful introduction rate but gave an almost linear reduction in colonized patient-days and ward-level prevalence. Shorter lengths of patient stay were accompanied by higher successful introduction rates, but had little effect on the other measures unless the mean time before detection of a colonized individual was large compared to the mean length of stay. We conclude that chance effects are likely to be amongst the most important factors in determining the course of an outbreak. Mathematical models can provide valuable insights into the non-linear interactions between a small number of processes, but for the very small populations found in hospital wards, a stochastic approach is essential.
针对诸如金黄色葡萄球菌等通过手部传播的医院病原体在普通内科-外科病房内的传播情况，构建了一个简单的数学模型。与先前的模型不同，该模型采用了随机方法。通过计算机模拟来探究模型的特性，并根据病原体的成功引入率、病房层面的患病率以及定植患者天数来呈现结果，着重强调患者及护理人员管理变化所产生的总体影响。病原体传播能力的微小变化会导致这三项指标都发生大幅变化。即使有效洗手频率的小幅增加也足以控制住地方性病菌。在广泛的不同情形下，减少入住病房的定植患者数量也是一种有效的控制措施。加强监测活动对成功引入率影响不大，但能使定植患者天数和病房层面的患病率几乎呈线性下降。患者住院时间较短伴随着较高的成功引入率，但对其他指标影响不大，除非与平均住院时间相比，发现定植个体之前的平均时间较长。我们得出结论，偶然因素很可能是决定疫情发展过程的最重要因素之一。数学模型能够为少数过程之间的非线性相互作用提供有价值的见解，但对于医院病房中数量极少的人群而言，采用随机方法至关重要。
