Department of Psychological Sciences, Texas Tech University, Lubbock, Texas, USA.
Department of Psychiatry, University of Rochester Medical Center, Rochester, New York, USA.
Suicide Life Threat Behav. 2021 Feb;51(1):76-87. doi: 10.1111/sltb.12670.
Categorical data analysis is relevant to suicide risk and prevention research that focuses on discrete outcomes (e.g., suicide attempt status). Unfortunately, results from these analyses are often misinterpreted and not presented in a clinically tangible manner. We aimed to address these issues and highlight the relevance and utility of categorical methods in suicide research and clinical assessment. Additionally, we introduce relevant basic machine learning methods concepts and address the distinct utility of the current methods.
We review relevant background concepts and pertinent issues with references to helpful resources. We also provide non-technical descriptions and tutorials of how to convey categorical statistical results (logistic regression, receiver operating characteristic [ROC] curves, area under the curve [AUC] statistics, clinical cutoff scores) for clinical context and more intuitive use.
We provide comprehensive examples, using simulated data, and interpret results. We also note important considerations for conducting and interpreting these analyses. We provide a walk-through demonstrating how to convert logistic regression estimates into predicted probability values, which is accompanied by Appendices demonstrating how to produce publication-ready figures in R and Microsoft Excel.
Improving the translation of statistical estimates to practical, clinically tangible information may narrow the divide between research and clinical practice.
分类数据分析与专注于离散结果（例如，自杀尝试状态）的自杀风险和预防研究相关。不幸的是，这些分析的结果经常被误解，并且没有以临床可感知的方式呈现。我们旨在解决这些问题，并强调分类方法在自杀研究和临床评估中的相关性和实用性。此外，我们介绍了相关的基本机器学习方法概念，并探讨了当前方法的独特用途。
我们回顾了相关的背景概念和参考有用资源的相关问题。我们还提供了非技术性的描述和教程，介绍如何为临床背景和更直观的用途传达分类统计结果（逻辑回归、接收者操作特征[ROC]曲线、曲线下面积[AUC]统计、临床截断分数）。
我们提供了全面的示例，使用模拟数据，并解释了结果。我们还注意到进行和解释这些分析的重要考虑因素。我们提供了一个演练，演示如何将逻辑回归估计值转换为预测概率值，附录中演示了如何在 R 和 Microsoft Excel 中生成可发表的图形。
提高统计估计值到实际、临床可感知信息的转化，可能会缩小研究和临床实践之间的差距。
