Department of Mathematics, University of Pavia, Pavia, Italy.
J Clin Monit Comput. 2010 Aug;24(4):269-77. doi: 10.1007/s10877-010-9244-2. Epub 2010 Jul 20.
To study the relationships between prevalence and incidence in the case of nosocomial infections of the urinary tract, and to evaluate if repeated prevalence measures may be useful to obtain an estimate of incidence.
Methodology is based on a simple and reasonable assumption on the infection dynamics: starting from a difference equation modeling the evolution of hospital population, it is obtained a set of equations allowing to calculate the incidence by means of the knowledge of prevalence.
The numerical validation of the model done by computer simulations, shows that the model obtains a better estimate of incidence than the approach given by the classical rule prevalence = incidence x duration.
The proposed strategy permits to forecast the incidence of the urinary tract nosocomial infections by using repeated measures of prevalence. It is hence possible to estimate the incidence from cross-sectional prevalence data with sufficient accuracy to monitor and estimate the time dynamics of these infections.
研究尿路感染的患病率和发病率之间的关系，并评估重复进行患病率测量是否有助于估计发病率。
该方法基于对感染动力学的简单合理假设：从一个简单的差分方程开始，该方程可对医院人群的演变进行建模，然后得到一组方程，通过患病率的知识可以计算发病率。
通过计算机模拟对模型进行了数值验证，结果表明，该模型比经典规则患病率=发病率×持续时间的方法能更好地估计发病率。
所提出的策略允许通过重复测量患病率来预测尿路感染的发病率。因此，可以使用横断面患病率数据来估计发病率，其准确性足以监测和估计这些感染的时间动态。
