著者

篠崎 智大 横田 勲 大庭 幸治 上妻 佳代子 坂巻 顕太郎

出版者

日本計量生物学会

雑誌

計量生物学 (ISSN:09184430)

巻号頁・発行日

vol.41, no.1, pp.1-35, 2020 (Released:2020-12-04)

参考文献数

65

---

Prediction models are usually developed through model-construction and validation. Especially for binary or time-to-event outcomes, the risk prediction models should be evaluated through several aspects of the accuracy of prediction. With unified algebraic notation, we present such evaluation measures for model validation from five statistical viewpoints that are frequently reported in medical literature: 1) Brier score for prediction error; 2) sensitivity, specificity, and C-index for discrimination; 3) calibration-in-the-large, calibration slope, and Hosmer-Lemeshow statistic for calibration; 4) net reclassification and integrated discrimination improvement indexes for reclassification; and 5) net benefit for clinical usefulness. Graphical representation such as a receiver operating characteristic curve, a calibration plot, or a decision curve helps researchers interpret these evaluation measures. The interrelationship between them is discussed, and their definitions and estimators are extended to time-to-event data suffering from outcome-censoring. We illustrate their calculation through example datasets with the SAS codes provided in the web appendix.

著者

田栗 正隆 高橋 邦彦 小向 翔 伊藤 ゆり 服部 聡 船渡川 伊久子 篠崎 智大 山本 倫生 林 賢一

出版者

日本計量生物学会

雑誌

計量生物学 (ISSN:09184430)

巻号頁・発行日

vol.44, no.2, pp.129-200, 2024 (Released:2024-04-25)

参考文献数

167

---

Epidemiology is the study of health-related states or events in specific populations and their determinants, with the aim of controlling health problems. It encompasses various research fields, such as cancer epidemiology, infectious disease epidemiology, and social epidemiology, molecular epidemiology, environmental epidemiology, genetic epidemiology, clinical epidemiology, pharmacoepidemiology, spatial epidemiology, and theoretical epidemiology, among others, and is closely related to statistics and biometrics. In analytical epidemiological studies, data is collected from study populations using appropriate study designs, and statistical methods are applied to understand disease occurrence and its causes, particularly establishing causal relationships between interventions or exposures and disease outcomes. This paper focuses on five topics in epidemiology, including infectious disease control through spatial epidemiology, cancer epidemiology using cancer registry data, research about long-term health effects, targeted learning in observational studies, and that in randomized controlled trials. This paper provides the latest insights from experts in each field and offers a prospect for the future development of quantitative methods in epidemiology.