Cummings S R, Richard R J
Division of General Internal Medicine, University of California, San Francisco 94143.
Am J Public Health. 1988 May;78(5):574-5. doi: 10.2105/ajph.78.5.574.
Selection of cutoff points for tests to validate smoking cessation should take account of the prevalence of deception. When the prevalence of deception is relatively low, the cutoff points to validate quitting should be relatively high. Many studies have used cutoff points that are too low and may have underestimated cessation rates. We present a method for determining the best cutoff points that takes account of the prevalence of deception.
用于验证戒烟的测试的临界值选择应考虑欺骗行为的发生率。当欺骗行为的发生率相对较低时,用于验证戒烟的临界值应相对较高。许多研究使用的临界值过低,可能低估了戒烟率。我们提出了一种考虑欺骗行为发生率来确定最佳临界值的方法。