Sheeran Paschal, Frisch Caroline, Listrom Olivia, Pei Yifei, Bermudez Andrea, Rothman Alexander J, Smith Jennifer S
Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill.
Department of Psychology, University of Minnesota.
Health Psychol. 2025 Apr;44(4):366-379. doi: 10.1037/hea0001444. Epub 2024 Dec 16.
The effectiveness of behavioral interventions is typically evaluated relative to control conditions using null hypothesis significance testing (i.e., < .05) or effect sizes. These criteria overlook comparisons with previous interventions and do little to promote a cumulative science of behavior change. We conducted a systematic review of the effectiveness of interventions to promote colorectal cancer screening (CCS) and generated benchmarks via the percentile distribution of s, screening rates for intervention and control arms, and differential screening rates (intervention minus control rate) in respective trials.
Literature searches identified 187 eligible tests ( = 371,018).
Random effects meta-analysis computed a sample-weighted = 1.69 (95% CI [1.55, 1.84]) and meta-regression showed that there was no improvement in the effectiveness of CCS interventions between 1996 and 2022. Benchmarking indicated that the median effect size was = 1.32, equivalent to a 35.7% screening rate in the intervention arm, and a 5.9% differential screening rate. Benchmarks were also generated for different types of screening (e.g., fecal immunochemical test, colonoscopy), sample characteristics (e.g., race, socioeconomic status), and methodological features (e.g., control conditions).
Interventions to promote CCS have a small effect and effectiveness has not increased over time. The percentile values for effect sizes and screening rates reported here can be used to benchmark the effectiveness of future trials. Benchmarking offers a way to evaluate interventions that are grounded in accumulated evidence and can inform judgments about tradeoffs among effectiveness, reach, and cost. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
行为干预的有效性通常通过零假设显著性检验（即p <.05）或效应量相对于对照条件进行评估。这些标准忽略了与先前干预措施的比较，对促进行为改变的累积科学几乎没有作用。我们对促进结直肠癌筛查（CCS）的干预措施的有效性进行了系统评价，并通过效应量的百分位数分布、干预组和对照组的筛查率以及各试验中的差异筛查率（干预组减去对照组的比率）生成了基准。
文献检索确定了187项符合条件的试验（N = 371,018）。
随机效应荟萃分析计算出样本加权效应量为d = 1.69（95%CI [1.55, 1.84]），荟萃回归显示1996年至2022年间CCS干预措施的有效性没有提高。基准分析表明，效应量的中位数为d = 1.32，相当于干预组35.7%的筛查率和5.9%的差异筛查率。还针对不同类型的筛查（如粪便免疫化学试验、结肠镜检查）、样本特征（如种族、社会经济地位）和方法学特征（如对照条件）生成了基准。
促进CCS的干预措施效果较小，且有效性并未随时间增加。此处报告的效应量和筛查率的百分位数值可用于衡量未来试验的有效性。基准分析提供了一种基于累积证据评估干预措施的方法，并可为有关有效性、覆盖面和成本之间权衡的判断提供参考。（PsycInfo数据库记录（c）2025美国心理学会，保留所有权利）
