P值应用思考：挑战与优化策略

doi:10.12092/j.issn.1009-2501.2026.02.011

中国临床药理学与治疗学 ›› 2026, Vol. 31 ›› Issue (2): 240-246.doi: 10.12092/j.issn.1009-2501.2026.02.011

P值应用思考：挑战与优化策略

刘志伟¹^,²(), 陈舒晴³(), 陈右君¹^,², 李一鸣¹^,², 言方荣⁴, 赵杨⁵, 孙华¹, 谢海棠¹^,*(), 王陵⁶^,*()

1. 皖南医学院弋矶山医院药物临床评价中心，芜湖 241001，安徽
2. 皖南医学院，芜湖 241002，安徽
3. 华中科技大学同济医学院，公共卫生学院，流行病学与生物统计学系，武汉 430030，湖北
4. 中国药科大学理学院，南京 211198，江苏
5. 南京医科大学公共卫生学院生物统计学系，南京 211166，江苏
6. 空军军医大学军事预防医学系卫生统计学教研室，西安 710032，陕西

收稿日期:2025-08-24 修回日期:2026-02-06 出版日期:2026-02-26 发布日期:2026-03-17
通讯作者: 谢海棠,王陵 E-mail:974973701@qq.com;632227690@qq.com;xiehaitang2023@163.com;lynnw@fmmu.edu.cn
作者简介:刘志伟，男，硕士研究生，研究方向：定量药理学。 E-mail：974973701@qq.com|陈舒晴，共同第一作者，女，博士研究生，研究方向：临床药学和统计学。E-mail：632227690@qq.com
基金资助:
2023年安徽省临床医学研究转化专项（202304295107020005）

Current application of P-value: challenges and optimization strategies

Zhiwei LIU¹^,²(), Shuqing CHEN³(), Youjun CHEN¹^,², Yiming LI¹^,², Fangrong YAN⁴, Yang ZHAO⁵, Hua SUN¹, Haitang XIE¹^,*(), Ling WANG⁶^,*()

1. Yijishan Hospital Clinical Trial Center, Wannan Medical College, Wuhu 241001, Anhui, China
2. Wannan Medical College, Wuhu 241002, Anhui, China
3. Department of Epidemiology and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, Hubei, China
4. School of Science, China Pharmaceutical University, Nanjing 211198, Jiangsu, China
5. Department of Biostatistics, School of Public Health, Nanjing Medical University, Nanjing 211166, Jiangsu, China
6. Department of Health Statistics, Faculty of Military Preventive Medicine, Air Force Medical University, Xi’an 710032, Shaanxi, China

Received:2025-08-24 Revised:2026-02-06 Online:2026-02-26 Published:2026-03-17
Contact: Haitang XIE,Ling WANG E-mail:974973701@qq.com;632227690@qq.com;xiehaitang2023@163.com;lynnw@fmmu.edu.cn

摘要/Abstract

摘要：

P值（基于频率学派）是统计学中用于检验原假设的重要工具，表示在原假设为真时观察到当前或更极端结果的概率。尽管P值被广泛用于判断研究结果的显著性（如P<0.05，P<0.01），但其滥用和误用也引发了诸多争议。本文综述了P值在科学研究与临床试验中的重要性及局限性，包括其对样本量的敏感性、二元分类的误导性，以及无法客观反映有无临床意义等问题。同时，探讨了作为补充和优化的方法，如置信区间、效应量、贝叶斯统计等。基于现有共识与争议，提出合理使用P值的建议，强调结合多种统计指标以提升科学推断的严谨性和可重复性，认识到P值在控制错误率和组间比较中的核心作用。未来，提倡将P值重新定位为原假设的“兼容性”指标，并与其他方法协同应用，避免简单的二元分类。

关键词: 统计学, P值, 研究进展, 统计滥用

Abstract:

The P-value (based on the frequentist school of thought) is an important tool in statistics for testing the null hypothesis, representing the probability of observing the current or more extreme results when the null hypothesis is true. Although P-values are widely used to determine the statistical significance of research findings (e.g., P<0.05, P<0.01), their abuse and misuse have sparked numerous controversies. This article reviews the importance and limitations of P-values in scientific research and clinical trials, including their sensitivity to sample size, the misleading nature of binary classification, and their inability to objectively reflect clinical significance, among other issues. Additionally, it explores supplementary and optimized methods, such as confidence intervals, effect sizes, and Bayesian statistics. Based on existing consensus and controversies, recommendations for the rational use of P-values are proposed, emphasizing the integration of multiple statistical indicators to enhance the rigor and reproducibility of scientific inferences, while recognizing the core role of P-values in controlling error rates and making group comparisons. Looking ahead, it advocates repositioning P-values as an indicator of the "compatibility" with the null hypothesis and applying them in conjunction with other methods to avoid simplistic binary classification.

Key words: statistics, P-value, research progress, statistical abuse

中图分类号:

C8
O212.1

刘志伟, 陈舒晴, 陈右君, 李一鸣, 言方荣, 赵杨, 孙华, 谢海棠, 王陵. P值应用思考：挑战与优化策略[J]. 中国临床药理学与治疗学, 2026, 31(2): 240-246.

Zhiwei LIU, Shuqing CHEN, Youjun CHEN, Yiming LI, Fangrong YAN, Yang ZHAO, Hua SUN, Haitang XIE, Ling WANG. Current application of P-value: challenges and optimization strategies[J]. Chinese Journal of Clinical Pharmacology and Therapeutics, 2026, 31(2): 240-246.

参考文献 50

1	Anonymous. Taking the P: why the founder of P-values would be turning in his grave [M]. London: Unknown Publisher, 2020: 1-50.
2	Fisher RA. Statistical methods for research workers [M]//Kotz S, Johnson NL. Breakthroughs in Statistics. New York: Springer, 1992: 66-70.
3	Dahiru T. P-Value, a true test of statistical significance? A cautionary note[J]. Ann Ibadan Postgrad Med, 2011, 6 (1): 21- 26. doi: 10.4314/aipm.v6i1.64038
4	McShane BB, Gal D. The ASA statement-related article: statistical significance and the dichotomization of evidence[J]. J Am Stat Assoc, 2017, 112 (519): 885- 895.
5	Hodzic-Santor E, Fralick M. Treating patients, not P values[J]. NEJM Evid, 2025, 4 (1): 1- 5.
6	Sharma H. Statistical significance or clinical significance? A researcher's dilemma for appropriate interpretation of research results[J]. Saudi J Anaesth, 2021, 15 (4): 431- 434. doi: 10.4103/sja.sja_158_21
7	Thiese MS, Ronna B, Ott U. P value interpretations and considerations[J]. J Thorac Dis, 2016, 8 (9): E928- E931. doi: 10.21037/jtd.2016.08.16
8	Rothman J. A show of confidence[J]. N Engl J Med, 1978, 299 (24): 1362- 1363.
9	Trafimow D. Editorial[J]. Basic Appl Soc Psychol, 2014, 36 (1): 1- 2. doi: 10.1080/01973533.2014.865505
10	Trafimow D, Marks M. Editorial[J]. Basic Appl Soc Psychol, 2015, 37 (1): 1- 2. doi: 10.1080/01973533.2015.1012991
11	Wasserstein RL, Lazar NA. The ASA statement on p-values: context, process, and purpose[J]. Am Stat, 2016, 70 (2): 129- 133. doi: 10.1080/00031305.2016.1154108
12	Harrington D, D'Agostino RB, Gatsonis C, et al. New guidelines for statistical reporting in the journal[J]. N Engl J Med, 2019, 381 (3): 285- 286. doi: 10.1056/NEJMe1906559
13	Young PJ, Nickson CP, Perner A. When should clinicians act on non-statistically significant results from clinical trials?[J]. JAMA, 2020, 323 (22): 2256- 2257. doi: 10.1001/jama.2020.3508
14	Ioannidis JPA. What have we (not) learnt from millions of scientific papers with P values?[J]. Am Stat, 2019, 73 (S1): 20- 25. doi: 10.1080/00031305.2018.1447512
15	Van Zwet E, Gelman A, Greenland S, et al. A new look at P values for randomized clinical trials[J]. NEJM Evid, 2023, 3 (1): 1- 8. doi: 10.1056/evidx2400007
16	Wasserstein RL, Schirm AL, Lazar NA. Moving to a world beyond "P < 0.05"[J]. Am Stat, 2019, 73 (S1): 1- 19.
17	Begg CB. In defense of P values[J]. JNCI Cancer Spectr, 2020, 4 (2): pkaa012. doi: 10.1093/jncics/pkaa012
18	Mayo DG, Hand D. Statistical significance and its critics: practicing damaging science, or damaging scientific practice?[J]. Synthese, 2022, 200 (3): 220. doi: 10.1007/s11229-022-03692-0
19	Lytsy P. P in the right place: revisiting the evidential value of P-values[J]. J Evid Based Med, 2018, 11 (4): 288- 291. doi: 10.1111/jebm.12319
20	Aguinis H, Vassar M, Wayant C. On reporting and interpreting statistical significance and P values in medical research[J]. BMJ Evid Based Med, 2021, 26 (2): 39- 42. doi: 10.1136/bmjebm-2019-111264
21	Ahmed ES, Butt MN. The misunderstood P-value: why statistical significance is not enough in clinical practice[J]. Br J Anaesth, 2025, 134 (4): 909- 913. doi: 10.1016/j.bja.2025.01.008
22	Stahel WA. New relevance and significance measures to replace P-values[J]. PLoS One, 2021, 16 (6): e0252991. doi: 10.1371/journal.pone.0252991
23	Gao J. P-values – a chronic conundrum[J]. BMC Med Res Methodol, 2020, 20 (1): 167. doi: 10.1186/s12874-020-01051-6
24	Polack FP, Thomas SJ, Kitchin N, et al. Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine[J]. N Engl J Med, 2020, 383 (27): 2603- 2615. doi: 10.1056/NEJMoa2034577
25	Lammers D, Richman J, Holcomb JB, et al. Use of Bayesian statistics to reanalyze data from the pragmatic randomized optimal platelet and plasma ratios trial[J]. JAMA Netw Open, 2023, 6 (2): e230421. doi: 10.1001/jamanetworkopen.2023.0421
26	Tran NA, McGrory A, Poonai N, et al. A comparison of alternative ranking methods in two-stage clinical trials with multiple interventions: an application to the anxiolysis for laceration repair in children trial [J]. Clin Trials, 2024, 21(5): 17407745241251812.
27	Manson JE, Cook NR, Lee IM, et al. Vitamin D supplements and prevention of cancer and cardiovascular disease[J]. N Engl J Med, 2019, 380 (1): 33- 44. doi: 10.1056/NEJMoa1809944
28	Manson JE, Cook NR, Lee IM, et al. Marine n−3 fatty acids and prevention of cardiovascular disease and cancer[J]. N Engl J Med, 2019, 380 (1): 23- 32. doi: 10.1056/NEJMoa1811403
29	Patel S, Green A. Death by p-value: the overreliance on p-values in critical care research[J]. Crit Care, 2025, 29 (1): 73. doi: 10.1186/s13054-025-05307-9
30	Marschner IC. Confidence distributions for treatment effects in clinical trials: posteriors without priors[J]. Stat Med, 2024, 43 (6): 1271- 1289. doi: 10.1002/sim.10000
31	Wilcox RR, Serang S. Hypothesis testing, P values, confidence intervals, measures of effect size, and Bayesian methods in light of modern robust techniques[J]. Educ Psychol Meas, 2017, 77 (4): 673- 689. doi: 10.1177/0013164416667983
32	Murtaugh PA. In defense of P values[J]. Ecology, 2014, 95 (3): 611- 617. doi: 10.1890/13-0590.1
33	Halsey LG. The reign of the P-value is over: what alternative analyses could we employ to fill the power vacuum?[J]. Biol Lett, 2019, 15 (5): 20190174. doi: 10.1098/rsbl.2019.0174
34	Homa-Bonell JK. Interpreting P values in 2023[J]. J Patient Cent Res Rev, 2023, 10 (3): 102- 103.
35	Amrhein V, Greenland S. Discuss practical importance of results based on interval estimates and P-value functions, not only on point estimates and null P-values[J]. J Inf Technol, 2022, 37 (3): 316- 320. doi: 10.1177/02683962221105904
36	Rovetta A, Piretta L, Mansournia MA. P-values and confidence intervals as compatibility measures: guidelines for interpreting statistical studies in clinical research[J]. Lancet Reg Health Southeast Asia, 2025, 33, 100534. doi: 10.1016/j.lansea.2025.100534
37	Freire APCF, Elkins MR, Ramos EMC, et al. Use of 95% confidence intervals in the reporting of between-group differences in randomized controlled trials: analysis of a representative sample of 200 physical therapy trials[J]. Braz J Phys Ther, 2019, 23 (4): 302- 310. doi: 10.1016/j.bjpt.2018.10.004
38	Fernandes AD, Vu MTHQ, Edward LM, et al. A reproducible effect size is more useful than an irreproducible hypothesis test to analyze high throughput sequencing datasets [EB/OL]. Cornell University: arXiv, 2019, arXiv: 1909.05208v1.
39	Valladares-Neto J. Effect size: a statistical basis for clinical practice[J]. Rev Odontoc Sci, 2018, 33 (1): 84. doi: 10.15448/1980-6523.2018.1.29437
40	Citrome L. Quantifying clinical relevance[J]. J Clin Psychopharmacol, 2020, 40 (2): 115- 121.
41	Storey JD. The positive false discovery rate: a Bayesian interpretation and the q-value[J]. Ann Stat, 2003, 31 (6): 2013- 2035. doi: 10.1214/aos/1074290335
42	Fithian W, Lei L. Conditional calibration for false discovery rate control under dependence[J]. Ann Stat, 2022, 50 (6): 3139- 3164.
43	Rafi Z, Greenland S. Semantic and cognitive tools to aid statistical science: replace confidence and significance by compatibility and surprise[J]. BMC Med Res Methodol, 2020, 20 (1): 244. doi: 10.1186/s12874-020-01105-9
44	Bai Z, Fujikoshi Y, Hu J. Strong consistency of the AIC, BIC, Cp and KOO methods in high-dimensional multivariate linear regression [EB/OL]. Cornell University: arXiv, 2020, arXiv: 2006.06152v1.
45	Rezaei M, Malekjani M. Comparison between different methods of model selection in cosmology[J]. Eur Phys J Plus, 2021, 136 (2): 219. doi: 10.1140/epjp/s13360-021-01200-w
46	Soukup P. Může Bayesův faktor nahradit P hodnotu?[J]. Sociální Stud, 2021, 18 (1): 73- 99. doi: 10.5817/SOC2021-1-73
47	Lu S, Liu K, Chen XY, et al. Bayesian methods as a complementary tool: balancing innovation and rigor in critical care research[J]. Crit Care, 2025, 29 (1): 135. doi: 10.1186/s13054-025-05380-0
48	Malone HE, Coyne I. Complementing the P-value from null-hypothesis significance testing with a Bayes factor from null-hypothesis Bayesian testing[J]. Nurse Res, 2020, 28 (4): 41- 48. doi: 10.7748/nr.2020.e1756
49	Kennedy-Shaffer L. When the alpha is the omega: P-values, "substantial evidence, " and the 0.05 standard at FDA [J]. Food Drug Law J, 2018, 73(4): 589-616.
50	Mayo D. P-values on trial: selective reporting of (best practice guides against) selective reporting[J]. Harv Data Sci Rev, 2020, 2 (3): 1- 18. doi: 10.1162/99608f92.e2473f6a

[1]	经加吻, 孟庆波, 毕正, 王帆竞, 李瑜璠, 方朝晖. 基于治疗方法的糖尿病勃起功能障碍动物模型研究进展[J]. 中国临床药理学与治疗学, 2025, 30(9): 1224-1232.
[2]	张译丹, 孙兴华, 曲阳, 胡晓阳, 张淼. 中药有效成分调控PI3K/Akt信号通路治疗脑缺血再灌注损伤的研究进展[J]. 中国临床药理学与治疗学, 2025, 30(6): 820-827.
[3]	丁丁, 吴胜男, 王安才, 王德国. 维立西呱抑制心脏重构的研究进展[J]. 中国临床药理学与治疗学, 2025, 30(6): 858-863.
[4]	刘叶元, 祁亚锋, 张茂福, 李欣钰, 沈雁云, 刘瑜, 张尚祖, 李洋洋, 张利英, 张志明. 中药干预化疗肾损伤研究进展[J]. 中国临床药理学与治疗学, 2025, 30(4): 556-569.
[5]	黄淑敏, 谢保城, 刘国辉. GLP-1受体激动剂治疗糖尿病肾病作用机制的研究进展[J]. 中国临床药理学与治疗学, 2025, 30(12): 1701-1710.
[6]	田杰祥, 漆文霞, 王钢, 闫彦峰, 汪湛东, 魏勇, 刘海龙, 张媛媛. 唾液腺上皮细胞在干燥综合征中的病理作用研究进展[J]. 中国临床药理学与治疗学, 2025, 30(11): 1550-1558.
[7]	罗加伟, 陈天旺, 王鑫宇, 刘娟, 黄波, 李利生, 徐尚福. 中药天然产物预防血管再狭窄的作用及机制研究进展[J]. 中国临床药理学与治疗学, 2025, 30(10): 1408-1416.
[8]	刘轩, 冯翠娟, 王一强, 李芳. miRNA在溃疡性结肠炎中的研究进展[J]. 中国临床药理学与治疗学, 2024, 29(8): 917-929.
[9]	余吉玲, 张晶, 邓毅凡, 何胜虎. ω-3 PUFA在心力衰竭中的研究进展[J]. 中国临床药理学与治疗学, 2024, 29(2): 236-240.
[10]	赵全铭, 杨满斗, 胡义波, 苏有橦, 普丽, 章瑜, 李文亮. 胃肠道间质瘤药物治疗和耐药机制研究进展[J]. 中国临床药理学与治疗学, 2024, 29(1): 82-89.
[11]	许金回, 岳红梅, 李雅亭, 刘苗苗, 武兴东, 朱浩斌. 间充质干细胞对OSAHS病理生理调节机制的研究进展[J]. 中国临床药理学与治疗学, 2024, 29(1): 114-120.
[12]	罗师萍, 张捷, 俞育帅, 宋传贵. HER2阳性乳腺癌的靶向治疗进展[J]. 中国临床药理学与治疗学, 2023, 28(8): 876-886.
[13]	张莉, 蔺兴遥, 尚芸, 王强. 冻伤的病理机制及治疗研究进展[J]. 中国临床药理学与治疗学, 2023, 28(3): 347-354.
[14]	李晨豪, 喻喜华, 闫文浩, 柴宇霞, 李燕燕, 李锟, 李泽运. 儿童癫痫与维生素D相关性研究进展[J]. 中国临床药理学与治疗学, 2023, 28(11): 1299-1306.
[15]	刘丹, 刘明, 金良友, 潘娟, 辛昊儒, 刘梦圆, 李鑫, 郑坤, 冯晓灵. 木香烃内酯抗肿瘤活性研究进展[J]. 中国临床药理学与治疗学, 2023, 28(10): 1168-1176.

P值应用思考：挑战与优化策略

Current application of P-value: challenges and optimization strategies

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献 50

相关文章 15

编辑推荐

Metrics

本文评价