临床研究中的样本量估算

摘要

临床研究确定样本量大小的基本原则是在保证研究结论具有一定可靠性、精确度、可重复性以及研究效率的前提下,估算出所需的最小样本例数。其对整个研究的科学性、伦理性和研究投入都带来重大影响。样本量估算既可通过公式计算,也可以查表得到。但二者都必须事先确定四个基本指标,它们也是样本量在不同设计类型估算公式推导的理论依据:1.检验水准(α),即第一类错误的概率。α值越小,所需样本量越大,α通常取0.05或0.01。2.检验效能(1-β),即把握度(power),β为第二类错误的概率。1-β的涵义是如果备择假设正确时,检验将发现它的能力。检验效能越大,所需样本量就越大。β通常取0.20或0.10。3.预期的效应大小(δ):研究者期望探测到的有临床意义的差异大小。这是唯一一个不由统计学,而由临床专业知识决定的参数,也是最容易被误解和用错的地方。一般预期的效应值越小,所需样本量越大。4.预期的变异大小:总体标准差(σ)或总体率(π)。当预期效应值固定时,如果计量数据σ越大,所需样本量越大；π越接近0.5,则所需样本量越大。一般情况下,检验水准取0.05、把握度最低取80%是大多数临床研究的标准设定,所以真正决定研究规模差异的是临床参数,这种估算可以从几种途径获得:①前人所作类似的研究资料；②研究做的一部分预试验；③在研究工作进行的过程中作出相应的估算。

参考文献

[1] ZHOU J, DANG S N, ZENG L X, et al. Rapid infancy weight gain and 7— to 9—year childhood obesity risk: a prospective cohort study in rural western China[J]. Medicine (Baltimore), 2016, 95(16): e3425.

[2] ZIEGLER E E. Nutrient needs for catch—up growth in low—birthweight infants[J]. Nestle Nutr Inst Workshop Ser, 2015, 81: 135-143.

[3] SIMMONDS M, LLEWELLYN A, OWEN C G, et al. Predicting adult obesity from childhood obesity: a systematic review and meta—analysis[J]. Obes Rev, 2016, 17(2): 95-107.

[4] GUO T, ZHENG S, CHEN T, et al. The association of long—term trajectories of BMI, its variability, and metabolic syndrome: a 30—year prospective cohort study[J]. E Clinical Medicine, 2024, 69: 102486.

[5] RAO W W, ZONG Q Q, ZHANG J W, et al. Obesity increases the risk of depression in children and adolescents: results from a systematic review and meta—analysis[J]. J Affect Disord, 2020, 267: 78-85.

[6] 阳秀英, 李新影. 青少年超重、肥胖与抑郁和行为问题的关系[J]. 中国心理卫生杂志, 2016, 30(7): 519-526.

[7] YANG X Y, LI X Y. Relations of overweight and obesity to depression and externalizing behavior in adolescents[J]. Chinese Mental Health Journal, 2016, 30(7): 519-526.

[8] KUNDI H, AMIN Z M, FRIEDMAN M, et al. Association of obesity with psychological distress in young adults: patterns by sex and race or ethnicity[J]. JACC Adv, 2024, 3(8): 101115.

[9] BHATNAGAR S, TANEJA S. Zinc and cognitive development[J]. Br J Nutr, 2001, 85 Suppl 2: S139-S145.

[10] BLACK M M. Micronutrient deficiencies and cognitive functioning[J]. J Nutr, 2003, 133(11 suppl 2): S3927-S3931.

[11] SCAGLIONI S, DE COSMI V, CIAPPOLINO V, et al. Factors influencing children's eating behaviours[J]. Nutrients, 2018, 10(6): 706.