Stephen C
Department of Health Care and Epidemiology, University of British Columbia, Vancouver, Canada.
Infect Control Hosp Epidemiol. 1996 Apr;17(4):262-6. doi: 10.1086/647290.
Medical researchers often are faced with the challenge of estimating the total number of cases in a population based on incomplete samples. Because of a lack of explicit methods for determining if all cases have been counted, indirect methods for estimating the abundance of disease have been developed. Capture-recapture models are an indirect method of estimating population sizes that have been employed in recent epidemiological studies. These methods, derived from techniques developed for studies of animal abundance, estimate the true population size by evaluating the degree of overlap among incomplete lists of cases from existing data sources. Although intuitively appealing, the successful application of these methods is dependent upon a clear understanding of the biology of the disorder involved, the dynamics of the reference population, and the assumption and robustness of the specific models used. Failure to address these issues can lead to inaccurate and sometimes misleading results. This article describes some of the strengths and limitations of recapture techniques and provides the reader with a foundation from which to explore the methods in further detail.
医学研究人员常常面临基于不完整样本估计总体病例数的挑战。由于缺乏确定是否已统计所有病例的明确方法，已开发出用于估计疾病流行程度的间接方法。捕获再捕获模型是一种估计总体规模的间接方法，已应用于近期的流行病学研究。这些方法源自为动物数量研究开发的技术，通过评估来自现有数据源的不完整病例列表之间的重叠程度来估计真实的总体规模。尽管这些方法直观上很有吸引力，但其成功应用取决于对所涉及疾病生物学、参考人群动态以及所用特定模型的假设和稳健性的清晰理解。未能解决这些问题可能导致不准确甚至有时会产生误导性的结果。本文描述了再捕获技术的一些优点和局限性，并为读者提供了进一步详细探索这些方法的基础。
