关于「实践与证伪方案:文案检查器、上架清单与预注册检验」,《釉质再矿化与口腔硝酸盐通路》给出的定量结论是什么?
结论:第八章给出攻心文案检查器、上架清单、客服红线与共需 1091 人的预注册检验方案。书中以标注框区分已发表的证据、可运行的 Python 计算(18 段程序)、原典与作者的假说。
数据口径:书中参数为示意;全部计算由书中程序生成;本书不是医疗或牙科建议,不宣称任何产品或做法的健康功效。 第二章说明临界 pH 不是常数,而是钙与磷的函数,在示意网格中从 5.12 变到 6.07,对钙浓度对数的斜率为 −0.722。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 1091, 18
| test n (alpha 0.05, power 0.9) | |
| T1 BP crossover (pairs) | 97 |
| T2 nitrite crossover (pairs) | 18 |
| T3 in situ slabs (per arm) | 87 |
| T4 non-inferiority (per arm) | 410 |
| T5 critical-pH slope (people) | 24 |
| people if T1, T3, T4 run separately: 1091 | |
查看完整验算程序
# Chapter 8. p82: sample sizes for the pre-registered tests of this book
# Normal-approximation formulas with a small-sample correction of
# +2 per group for t-tests (Guenther's rule of thumb).
# T1 crossover, mouthwash vs water rinse, home systolic pressure:
# detect 2.0 mmHg, SD of within-person difference 6 mmHg.
# T2 same crossover, log plasma nitrite: detect a 25 % fall,
# SD of within-person log difference 0.35.
# T3 nano-HAP vs fluoride paste, in situ enamel slabs, mineral gain:
# standardised difference d = 0.5, two arms.
# T4 non-inferiority on caries increment, p = 0.735, margin 0.10.
# T5 critical-pH test: slope of critical pH on log10 calcium, true
# -0.72 (p22) against a null of 0; residual SD 0.15 pH units,
# log10 Ca spread SD 0.15 across volunteers.
# All SDs are planning assumptions, to be replaced by pilot data.
import math
from statistics import NormalDist
Z = NormalDist().inv_cdf
ZA2, ZA1, ZB = Z(0.975), Z(0.975), Z(0.90)
def paired(delta, sd):
return math.ceil(((ZA2 + ZB) * sd / delta) ** 2) + 2
def two_arm(d):
return math.ceil(2 * ((ZA2 + ZB) / d) ** 2) + 2
def noninf(p, margin):
return math.ceil((ZA1 + ZB) ** 2 * 2 * p * (1 - p) / margin ** 2)
def slope_n(beta, sd_res, sd_x):
return math.ceil(((ZA2 + ZB) * sd_res / (abs(beta) * sd_x)) ** 2) + 3
if __name__ == "__main__":
rows = [
("T1 BP crossover (pairs)", paired(2.0, 6.0)),
("T2 nitrite crossover (pairs)", paired(-math.log(0.75), 0.35)),
("T3 in situ slabs (per arm)", two_arm(0.5)),
("T4 non-inferiority (per arm)", noninf(0.735, 0.10)),
("T5 critical-pH slope (people)", slope_n(-0.72, 0.15, 0.15)),
]
print("test n (alpha 0.05, power 0.9)")
for name, n in rows:
print("{:30} {:5d}".format(name, n))
total = math.fsum([rows[0][1], 2 * rows[2][1], 2 * rows[3][1]])
print("people if T1, T3, T4 run separately: {}".format(int(total)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《纳米羟基磷灰石溶度积成核、釉质微孔扩散修复与口腔共生菌硝酸盐还原通路》第 8 章「实践与证伪方案:文案检查器、上架清单与预注册检验」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-165 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026165,
author = {Fan, Yafu},
title = {Solubility-Product Nucleation of Nano-Hydroxyapatite, Enamel Micropore Diffusion and the Oral Commensal Nitrate-Reduction Pathway},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}