外气疗效:个体依赖性与重复性的科学检验

Written by

in

Language

Evidence level: A

外气疗法的”疗效”更像一把时灵时不灵的钥匙——换个人、换个疗程、换个实验室,结果就变;这不是玄学式的否定,而是当前证据最诚实的样貌。

正文

“外气”(External Qigong)指气功师不发触体接触、隔空向患者”发功”以治病的方式,是气功理论里最神秘、也最容易被质疑的一环。现代医学对它做过什么?答案出乎意料:不但有随机对照试验(RCT),还有专门的系统综述和荟萃分析。而它们给出的共同结论,比”有效”或”无效”更耐人寻味——疗效高度依赖”谁在发功”,且极难稳定重复。

最经典的证据来自一项膝骨关节炎盲法 RCT:112 名患者被随机分到外气组或假治疗组(演员模仿发功动作,患者与评估医生均不知分组),由两位气功师各自独立完成治疗。结果两位气功师的效果截然不同:气功师 2 的患者疼痛改善(-25.7 vs 假治疗 -13.1,p<0.01)和功能改善都显著优于对照,而气功师 1 的患者与对照组没有差异(PMID 18654733)。研究者据此明确写道:”并非所有外气治疗师都是等价的”,疗效似乎取决于治疗师的某种品质。这是”个体依赖性”最干净的一次呈现——实验条件完全相同,变的只有人。

个体依赖的另一面是”经验梯度”。Cochrane 对触摸疗法(Healing Touch、Therapeutic Touch、Reiki)的荟萃分析纳入 24 项研究、1153 名受试者,发现触摸组疼痛平均降低 0.83 单位(0-10 分尺,95%CI -1.16 ~ -0.50),且”由经验更丰富的实践者实施的试验,止痛效果更大”(PMID 18843720)。Reiki 的专项荟萃同样得到显著但脆弱的效应(4 项 RCT、212 人,SMD -0.927;PMID 29551623)。Jain 对 66 项生物场疗法研究的系统综述则给出更谨慎的分级:疼痛人群证据较强,癌症疲劳、生活质量等结局证据”模棱两可”,总体研究质量仅属中等(PMID 19856109)。

如果说临床层面还能用”个体差异”解释,那么体外实验就把重复性问题摆上了台面。加州太平洋医学中心的团队用严格双盲设计,让气功师隔着至少 10 厘米对培养的人脑细胞”发功”20 分钟,观察细胞克隆形成率:预实验(8 次实验)呈趋势(P=0.162),正式实验(28 次)出现显著增殖(P=0.036),然而复制实验(60 次)完全阴性(P=0.465)(PMID 15102336)。同一个团队、同一套方案,从显著到阴性的完整轨迹,正是”首阳后复制失败”的教科书案例。注意,三阶段实验都用了随机化与盲法,排除了”设计不严谨”的简单解释——复制实验的阴性是真阴性,而非方法失误。再加上一项对 21 项治疗性触摸研究的快速证据评估发现,18 项报告阳性结果但只有 4 项偏倚风险较低,”没有找到任何被声称益处的高质量证据”(PMID 33742792)——方法学质量本身也成了疗效可信度的绊脚石。

为什么”人”的因素如此关键?把外气研究放进更大图景就能看清:针灸、心理治疗、甚至药物临床试验里,”治疗者效应”(therapist effect)都是公认的混杂源,但常规研究设计默认治疗者之间是同质的,很少单独建模。外气研究的特殊性在于——治疗手段本身没有可标准化的”给药剂量”,疗效与治疗师几乎不可分离。于是同样的研究,换个治疗师就可能从阳性翻成阴性;这解释了文献里大量互相矛盾的结果,也解释了为什么荟萃分析只能得出”可能有效但证据薄弱”的折中结论。对科学而言,”效果依赖谁来做”并不等于”效果不存在”,而是意味着:在找到可测量的”有效成分”之前,外气的疗效无法被当作一个稳定可复制的医学事实来对待。

那么,这是否意味着外气”不存在”?不是。更准确的读法是:效应可能存在,但幅度小、个体差异大、机制未明,以至于常规统计手段难以稳定捕捉。慢性疼痛 RCT 显示外气在第 2-4 周显著降痛、但 8 周随访时差异不再显著(PMID 20626055),暗示效应还会随时间衰减。这些特征(小效应、异质性、衰减)恰恰指向一个可操作的科研议程:与其争论”有没有效”,不如把”治疗师特质”标准化、把效应量异质性量化、用多中心盲法 RCT 检验重复性——这也正是气功研究领域自己的呼吁。

证据与局限

已验证: 同一盲法 RCT 中,不同气功师的疗效存在显著差异(PMID 18654733);经验更丰富的实践者疗效更大(Cochrane 荟萃,PMID 18843720)。
部分验证: 外气对慢性疼痛的短期(4 周内)止痛效应在多篇 RCT 中可重复,但随访普遍衰减(PMID 20626055);体外细胞增殖效应”首阳后复制失败”(PMID 15102336)。
待验证: 治疗师”特质”能否被客观标准化测量(修为等级×疗效的剂量-反应);效应量异质性在多大程度上可被治疗师、受试者、疗程等变量解释;多中心盲法 RCT 能否稳定复现外气效应。

相关阅读

证据等级: A
参考文献: PMID 18654733, PMID 18843720, PMID 20626055, PMID 19856109, PMID 29551623, PMID 15102336, PMID 33742792


免责说明:本文属传统知识与科学文献的整理、对照与研究设计讨论,仅供理解与参考,不构成医疗建议,不能替代专业诊断与治疗。文中涉及的效应量、频率区间与暴露结论多为研究报告层面的推断,个体情况差异很大;如有身体不适或健康疑虑,请就医并由专业医师判断。

Scientific Validation of Qi Therapy: Individual Dependency and Reproducibility

Evidence level: A

The therapeutic efficacy of external qi therapy resembles a key that works intermittently—results vary depending on the practitioner, the treatment course, or the laboratory setting. This is not a metaphysical denial, but rather the most honest representation of current evidence.

Main Text

“External Qi” (External Qigong) refers to the practice where a qigong master directs energy toward a patient through the air without physical contact to treat illness. It is the most mysterious and frequently questioned component within qigong theory. Modern medicine has subjected it to rigorous scrutiny, yielding results that are surprising: not only Randomized Controlled Trials (RCTs) but also dedicated systematic reviews and meta-analyses exist. Their shared conclusion is more nuanced than simply “effective” or “ineffective”—efficacy depends heavily on “who is performing the therapy,” and it is extremely difficult to replicate stably.

The most classic evidence comes from a blinded RCT on knee osteoarthritis: 112 patients were randomly assigned to an external qi group or a sham treatment group (actors mimicking the energy emission gestures, with both patients and assessing physicians unaware of group allocation), treated independently by two qigong masters. The results showed stark differences between the two masters: patients treated by Master 2 demonstrated significant improvements in pain reduction (-25.7 vs. sham -13.1, p<0.01) and function compared to controls, whereas patients treated by Master 1 showed no difference from the control group (PMID 18654733). The researchers explicitly noted: “Not all external qi therapists are equivalent,” suggesting efficacy depends on a specific quality of the therapist. This represents the clearest presentation of individual dependency—experimental conditions were identical, varying only by the person.

The other side of individual dependency is the “experience gradient.” A Cochrane meta-analysis on healing touch therapies (Healing Touch, Therapeutic Touch, Reiki) included 24 studies with 1,153 participants, finding that the touch group showed an average pain reduction of 0.83 units (0-10 scale, 95%CI -1.16 ~ -0.50), with trials conducted by practitioners with more experience showing greater analgesic effects (PMID 18843720). A dedicated meta-analysis for Reiki similarly yielded a significant but fragile effect (4 RCTs, 212 participants, SMD -0.927; PMID 29551623). Jain’s systematic review of 66 biofield therapy studies offered a more cautious grading: evidence for pain populations was stronger, while outcomes such as cancer fatigue and quality of life were “equivocal,” with overall study quality rated as moderate (PMID 19856109).

If clinical levels can be explained by “individual differences,” in vitro experiments bring the reproducibility issue to the forefront. A team at the Pacific Medical Center in California used a strict double-blind design, having qigong masters direct energy toward cultured human brain cells across a gap of at least 10 cm for 20 minutes, observing cell clone formation rates: the pilot study (8 experiments) showed a trend (P=0.162), the formal experiment (28 experiments) showed significant proliferation (P=0.036), yet the replication experiment (60 experiments) was completely negative (P=0.465) (PMID 15102336). The same team, using the same protocol, exhibited a complete trajectory from significant to negative results, serving as a textbook case of “positive initial findings followed by replication failure.” Note that all three stages utilized randomization and blinding, excluding simple explanations of “poor study design”—the negative result in the replication phase was a true negative, not a methodological error. Furthermore, a rapid evidence assessment of 21 therapeutic touch studies found that while 18 reported positive results, only 4 had low bias risk, finding “no high-quality evidence for any claimed benefits” (PMID 33742792)—methodological quality itself became an obstacle to the credibility of efficacy claims.

Why is the human factor so critical? Placing external qi studies within a broader context reveals that the “therapist effect” is an acknowledged confounding source in acupuncture, psychotherapy, and even drug trials, yet conventional study designs typically assume homogeneity among therapists and rarely model them separately. The uniqueness of external qi research lies in the fact that the therapeutic modality itself lacks a standardized “dosage,” making efficacy inseparable from the therapist. Consequently, the same study could yield positive or negative results depending on the practitioner; this explains the numerous contradictory findings in the literature and why meta-analyses can only arrive at a compromise conclusion of “possibly effective but weak evidence.” For science, “efficacy depends on who performs it” does not equate to “efficacy does not exist,” but rather implies that until a measurable “active component” is identified, the efficacy of external qi cannot be treated as a stable, replicable medical fact.

So, does this mean external qi “does not exist”? No. A more accurate reading is: the effect may exist, but it is small in magnitude, highly heterogeneous among individuals, and its mechanism remains unclear, making it difficult for conventional statistical methods to capture stably. Chronic pain RCTs show external qi significantly reduces pain in weeks 2-4 but shows no significant difference at the 8-week follow-up (PMID 20626055), suggesting the effect may attenuate over time. These characteristics (small effect size, heterogeneity, attenuation) point toward a viable research agenda: rather than debating “whether it works,” focus on standardizing therapist traits, quantifying effect size heterogeneity, and validating reproducibility through multi-center blinded RCTs—precisely what the qigong research community itself has advocated.

Evidence and Limitations

Verified: Significant efficacy differences exist between different qigong masters within the same blinded RCT (PMID 18654733); practitioners with more experience show greater efficacy (Cochrane meta-analysis, PMID 18843720).
Partially Verified: Short-term (within 4 weeks) analgesic effects of external qi on chronic pain are reproducible across multiple RCTs, but follow-up generally shows attenuation (PMID 20626055); in vitro cell proliferation effects show “positive initial findings followed by replication failure” (PMID 15102336).
To Be Verified: Whether therapist “traits” can be objectively standardized (e.g., cultivation level × efficacy dose-response); to what extent effect size heterogeneity can be explained by variables such as therapist, subject, or course duration; whether multi-center blinded RCTs can stably reproduce external qi effects.

Related Reading

Evidence Level: A
References: PMID 18654733, PMID 18843720, PMID 20626055, PMID 19856109, PMID 29551623, PMID 15102336, PMID 33742792


Disclaimer: This article compiles traditional knowledge and scientific literature for comparison and research-design discussion only. It is not medical advice and cannot replace professional diagnosis or treatment. Effect sizes, frequency ranges and exposure conclusions cited here are largely inferences at the research-report level, and individual circumstances vary widely. If you have any discomfort or health concerns, please consult a physician.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *