关于「反安慰剂:同一条公式的负号」,《开放标签与价格标签》给出的定量结论是什么?
结论:第五章复算他汀的 n-of-1 试验,反安慰剂比为 0.892,他汀月份的症状负担中只有 0.055 来自药物本身,并建立警示标签的误归因模型。
数据口径:书中参数为示意;全部计算由书中程序生成;本书不是医疗建议,不宣称任何产品或做法对身体或健康有任何作用,书中流程与工具为非个人化经营建议。 书中以标注框区分已发表证据、可运行的 Python 计算(21 段程序)、原典与作者假说。
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: 0.055, 0.892
| no tablet mean | 8.0 SE from CI 1.68 |
| placebo mean 15.4 SE from CI 1.68 | |
| statin mean 16.3 SE from CI 1.68 | |
| nocebo ratio from group means: (15.4 - 8.0) / (16.3 - 8.0) | 0.892 |
| published pooled ratio 0.90; per-patient ratio 2.2 was unstable | |
| statin-month burden split: | |
| present without any tablet | 0.491 |
| added by taking a tablet | 0.454 |
| added by the statin itself | 0.055 |
| restarted within six months | 30/60 = 0.50 |
| completed all 12 months | 49/60 = 0.817 |
查看完整验算程序
# Chapter 5. p51: The statin n-of-1 trial and the nocebo ratio.
# Published inputs: Wood et al. (2020), NEJM letter: 60 patients
# randomised, 49 completed 12 months; mean symptom score (0-100) 8.0
# (95% CI 4.7 to 11.3) in no-tablet months, 15.4 (12.1 to 18.7) on
# placebo, 16.3 (13.0 to 19.6) on atorvastatin; statin vs placebo
# P = 0.39; pooled nocebo ratio 0.90; original per-patient ratio 2.2
# (CI -62.3 to 66.7); 30 of 60 restarted statins within six months.
# The ratio of group means and the SEs from the CIs are ours.
# 他汀 n-of-1 试验:反安慰剂比的两种算法与症状负担的拆分。
from statistics import NormalDist
Z = NormalDist()
Z975 = Z.inv_cdf(0.975)
MONTHS = {"no tablet": (8.0, 4.7, 11.3), "placebo": (15.4, 12.1, 18.7),
"statin": (16.3, 13.0, 19.6)}
none, plac, stat = (MONTHS[k][0] for k in ("no tablet", "placebo",
"statin"))
for k, (m, lo, hi) in MONTHS.items():
print("{:<9} mean {:4.1f} SE from CI {:.2f}".format(
k, m, (hi - lo) / (2 * Z975)))
ratio = (plac - none) / (stat - none)
print("nocebo ratio from group means: ({:.1f} - {:.1f}) / ({:.1f} - {:.1f})"
" = {:.3f}".format(plac, none, stat, none, ratio))
print("published pooled ratio 0.90; per-patient ratio 2.2 was unstable")
print("statin-month burden split:")
print(" present without any tablet {:.3f}".format(none / stat))
print(" added by taking a tablet {:.3f}".format((plac - none) / stat))
print(" added by the statin itself {:.3f}".format((stat - plac) / stat))
print("restarted within six months: {}/{} = {:.2f}".format(30, 60, 0.5))
print("completed all 12 months: {}/{} = {:.3f}".format(49, 60, 49 / 60))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《开放标签安慰剂预测编码、价格标签期望增益与杯弓蛇影反安慰剂母典》第 5 章「反安慰剂:同一条公式的负号」(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-201 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026201,
author = {Fan, Yafu},
title = {Open-Label Placebos and Predictive Coding, Price-Tag Expectancy Gains and the Nocebo of the Bow in the Cup},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}