DOI QR코드

DOI QR Code

진단검사에서 측도들의 효과적인 표현: 그래프를 활용한 방법

The Effect Measures for Diagnostic Test: A Graph Approach

  • 조태경 (동국대학교 정보통계학과) ;
  • 손창균 (동국대학교 정보통계학과)
  • Cho, Tae-Kyoung (Department of Information & Statistics Science, Dongguk University) ;
  • Son, Chang-Kyoon (Department of Information & Statistics Science, Dongguk University)
  • 투고 : 2012.10.15
  • 심사 : 2012.11.27
  • 발행 : 2013.02.28

초록

임상연구나 역학조사에서 $2{\times}2$ 분할표는 요약통계를 표현하는데 매우 유용하게 활용된다. 이때 $2{\times}2$ 분할표의 셀값들은 위험비, 상대위험비, 또는 승산비 등과 같은 척도를 계산하는데 이용된다. 본 논문에서는 민감도, 특이도, 예측도, 우도비와 같은 진단 측도들 간의 관계를 쉽게 이해할 수 있도록 MS-office 액셀을 이용한 방사형 그래프를 통해 시각화하는 개선된 방법을 제안하였다. 몇 가지 수치적 예제를 활용하여 제안된 방법의 활용성을 제시하였다.

In clinical study or epidemic research, the $2{\times}2$ frequency table is useful to present a summary statistic The values of four cells in $2{\times}2$ table use to calculate the effect measures such as risk ratio, relative ris ratio or odds ratio. In this paper, we suggest that the improved visualization method using a radar diagram supported by MS-office Excel from the $2{\times}2$ frequency table is able to understand and draw easily betweendiagnosti measures such as sensitivity, specificity, predictivity, and likelihood ratio. We use some numerical example in order to show the usage of the proposed method.

키워드

참고문헌

  1. Bradly, G. W. (1993). Disease, Diagonstic, and Decisions, New York, Wiley.
  2. Johnson, K. M. (1999). The two by two diagram: A graphical truth table, Journal of Clinical Epidemiology, 52, 1073-1082. https://doi.org/10.1016/S0895-4356(99)00087-6
  3. Kraemer, H. C. (2004). Reconsidering the odds ratio as measure of $2{\times}2$ association in a population, Statistics in Medicine, 23, 257-270. https://doi.org/10.1002/sim.1714
  4. Linn, S. (2005). New patient-oriented diagnostic test characteristics analogous to the likelihood ratios conveyed information on trustworthiness, Journal of Clinical Epidemiology, 58, 450-457. https://doi.org/10.1016/j.jclinepi.2004.07.009
  5. Michael, P. (2007). Excel 2007 in Easy Steps, In Easy Steps Limited.
  6. Ransohoff, D. F. (2002). Challenges and oppertunities in evaluating diagnostic tests, Journal of Clinical Epidemiology, 55, 1178-1182. https://doi.org/10.1016/S0895-4356(02)00524-3
  7. Sackett, D. L., Haynes, R. B., Guyatt, G. H. and Tugwell, P. (1991). Clinical Epidemiology, 2nd ed. Boston, MA: Little Brown.
  8. Walter, S. D. (2000). Choice of effect measure for epidemiological data, Journal of Clinical Epidemiology, 53, 931-939. https://doi.org/10.1016/S0895-4356(00)00210-9