Evidence level: A
“正气存内,邪不可干”在流行病学里有一个精确的统计对应物——效应修饰:同一个太阳周期暴露,基线维生素D不足的人反应强烈,充足的人几乎免疫。这个交互项,可能就是”虚则受邪”的现代表达。
正文
上一篇文章(血清维生素D:太阳周期影响健康的中介通道)提出了一个问题:太阳周期→疾病流行这条链里,有多少份额经 25(OH)D 传导?中介分析问的是”走哪条路”;本文要问的是另一个正交的问题——同样暴露在太阳周期里,谁受影响的强度更大? 统计上这叫效应修饰(effect modification),临床上有个更直白的说法:基线状态决定易感性。
先看干预端的铁证:补同样的维生素D,效果完全取决于你缺不缺。 2017 年 BMJ 发表的个体参与者数据 meta 分析(25 个 RCT、10,933 人)是这个问题上迄今最权威的证据:基线 25(OH)D 低于 25 nmol/L 的人,补充维生素D使急性呼吸道感染风险降低 42%(OR 0.58,需治数 8);而基线水平 ≥25 nmol/L 的人,补充后没有任何显著获益(OR 0.89,95%CI 0.77-1.04)(PMID 28202713)。这不是剂量问题,是”地基”问题——同样一剂药,缺的人是在”补漏”,不缺的人是在”加高”。
2024 年的新 RCT 把这张图补完整了。 一项在芬兰年轻男性中做的随机对照试验 + 14 周随访研究给出一个看似矛盾的组合:补充组与安慰剂组在急性呼吸道感染结局上无差异(每天 20µg 维生素D3 没能压低感染率),但无论分组,基线 25(OH)D 不足(<50 nmol/L)者的感染风险是充足者的 2.1 倍(OR 2.1,95%CI 1.2-3.7),缺勤天数 2.3 倍,且抗菌肽 cathelicidin 浓度显著更低(OR 0.49)(PMID 39301110)。读起来矛盾,其实是同一个故事的两面:短期补充救不回一个早已建立的基线状态,但基线状态本身就是最强的风险标记。2021 年《柳叶刀-糖尿病与内分泌学》汇总 46 个 RCT、75,541 人的结果与此一致——总体保护效应 8%(OR 0.92),而每日小剂量方案(400-1000 IU)的亚组保护最强(OR 0.70)(PMID 33798465)。
观察端同样指向”基线分层”而非”人人平等”。 英国 1958 年出生队列 6,789 人显示,呼吸道感染的流行有强季节性,且与血清 25(OH)D 的季节曲线严格反向:25(OH)D 每升高 10 nmol/L,感染风险降低 7%(PMID 21736791)。奥地利一个 655 人的危重病研究用了一个更聪明的做法——按月份特异的 25(OH)D 三分位分层(把季节波动本身当基线),低三分位组医院死亡率是高三分位组的 2.05 倍(PMID 24661739)。这两篇告诉我们:所谓”季节性”,本质上可能就是”冬天里缺乏的人暴露于冬天的病毒”的叠加;季节不是暴露本身,而是暴露的上游驱动器。
这个假说不是新提的——它最早来自 1981 年 Hope-Simpson 的”季节性刺激”假说。 这位流行病学家在研究流感大流行多年后提出:一个与太阳辐射紧密相关的”季节性刺激”(seasonal stimulus)解释了流感的季节爆发。2006 年的一篇评论把这条线收拢:太阳辐射驱动皮肤合成维生素D,冬季缺乏普遍,而 1,25(OH)2D 作为类固醇激素深度调节免疫——刺激抗菌肽表达、压制过度炎症因子、增强巨噬细胞氧化爆发;志愿者在冬季接种减毒流感病毒更容易发热和产生血清学反应;补充维生素D能降低儿童呼吸道感染(PMID 16959053)。注意这个假说的结构:它说的不是”太阳辐射直接杀病毒”,而是”太阳辐射通过维生素D改变宿主防线”——换句话说,流感季节性的宿主端解释,本来就是”缺乏者受邪”模型。五运六气的”时气致病”,在分子水平上可能就是这句话。
这正好接上五运六气的”虚则受邪”。 中医讲”邪之所凑,其气必虚”——同一个时令之气(太阳周期、气候节律),对正气充足者是”四时之气,各化其化”,对正气虚者是”虚邪贼风,避之有时”都避不开。用现代语言翻译:环境暴露的效应量不是常数,是基线状态的函数。太阳周期→紫外线→维生素D这条链上,”缺乏者敏感、充足者免疫”已经在 RCT 层面被反复证实(PMID 28202713/39301110),那么一个自然的推论是:太阳活动年(UVB 通量低、冬季更长)对健康结局的影响,应当集中在基线 25(OH)D 缺乏的人群里——这是可以在现有队列里直接检验的假设。
这个假设为什么值得单独建个课题? 因为它的统计意义被中介分析掩盖了。中介分析回答”多少效应经维生素D传导”,隐含假设是全人群同质;而效应修饰回答”这个传导是否只对部分人成立”,允许人群异质。二者可以共存:中介比例可能整体不高(因为存在非维生素D通路,见 uv-pathway-vitamin-d-dependent-independent),但缺乏亚组里的中介比例可能很高——这恰恰是”太阳周期-疾病”关联过去几十年测来测去时有时无的原因之一:不同研究里缺乏者的占比不同,效应量就不同。把这个交互项显式建模,可能是解开太阳-健康关联异质性的一把钥匙。
证据与局限
已验证: 基线 25(OH)D <25 nmol/L 者补充维生素D显著降低急性呼吸道感染(OR 0.58),≥25 nmol/L 者无获益(PMID 28202713);基线不足(<50 nmol/L)者感染风险 OR 2.1、cathelicidin 更低,且短期补充无法逆转(PMID 39301110);25(OH)D 与感染风险线性负相关、季节曲线反向(PMID 21736791);按季节特异三分位分层的低水平者危重病死亡 HR 2.05(PMID 24661739);补充剂总体 OR 0.92、日服方案亚组 OR 0.70(PMID 33798465);”季节性刺激”假说框架成立,冬季补充维生素D可降低儿童呼吸道感染(PMID 16959053)。
部分验证: “太阳周期-疾病”关联存在但效应量不稳(见 sunspot-latitude-health-effect-comparison),”缺乏者占比导致效应量漂移”是解释异质性的候选机制之一;冬季/高纬缺乏风险由纬度、肤色、GC 基因型(rs4588)共同决定(PMID 33230496),为”谁更容易受太阳周期影响”提供了可遗传的分层变量。
待验证: 太阳周期相位(黑子数/UVB 通量)× 基线 25(OH)D 的显式交互项建模(缺乏亚组中”太阳-疾病”关联是否显著更强);效应修饰与中介比例的联合分解(缺乏亚组的中介份额是否高于全人群);跨纬度梯度(高纬缺乏率高 → 太阳效应应更强);GC rs4588 基因型作为”基线状态”工具变量的孟德尔随机化设计。
相关阅读
- 血清维生素D:太阳周期影响健康的中介通道有多大分量?
- 太阳周期效应中”维生素D依赖 vs 非依赖”两条UV通路贡献的拆分检验
- 太阳活动效应的跨纬度多中心直接对比
- 五运六气=气候周期医学
- 研究池子:环境-生命频率接口
证据等级: A(效应修饰本身有 PubMed 直接支撑:IPD meta 亚组分析 PMID 28202713、RCT 基线分层 PMID 39301110;”太阳周期×基线状态交互项”尚无直接检验,属延伸推论)
参考文献: PMID 28202713, PMID 39301110, PMID 33798465, PMID 21736791, PMID 24661739, PMID 16959053, PMID 33230496
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。
Why do some people feel the effects of the same sun while others don’t? The effect modification of baseline vitamin D status
Evidence level: A
“The internal vital energy is intact, and the pathogen cannot invade” has a precise statistical counterpart in epidemiology known as effect modification: identical solar cycle exposure elicits a strong response in individuals with insufficient baseline Vitamin D, while those with sufficient levels remain largely immune. This interaction term may represent the modern expression of “the deficient suffer from the pathogen.”
Main Text
The previous article (Serum Vitamin D: The Mediating Role of the Solar Cycle in Health) posed a question: within the chain Solar Cycle → Disease Epidemic, what proportion of the effect is mediated by 25(OH)D? Mediation analysis asks “which path is taken”; this article addresses a different orthogonal question: Who experiences greater intensity of impact given identical exposure to the solar cycle? Statistically, this is termed effect modification; clinically, it is more plainly described as baseline status determining susceptibility.
First, look at the ironclad evidence from the intervention side: supplementing the same amount of Vitamin D yields results entirely dependent on deficiency. The 2017 individual participant data meta-analysis published in the BMJ (25 RCTs, 10,933 participants) remains the most authoritative evidence on this issue: individuals with baseline 25(OH)D <25 nmol/L experienced a 42% reduction in acute respiratory infection risk with Vitamin D supplementation (OR 0.58, NNT 8); conversely, those with baseline levels ≥25 nmol/L showed no significant benefit (OR 0.89, 95%CI 0.77-1.04) (PMID 28202713). This is not a dosage issue but a “foundation” issue—the same dose fills a leak for the deficient and raises a wall for the sufficient.
A 2024 RCT completes this picture. A randomized controlled trial with 14-week follow-up conducted in young Finnish men yielded a seemingly contradictory combination: no difference in acute respiratory infection outcomes between the supplementation and placebo groups (20µg Vitamin D3 daily failed to suppress infection rates), yet regardless of group assignment, individuals with baseline 25(OH)D insufficiency (<50 nmol/L) had an infection risk 2.1 times higher than the sufficient group (OR 2.1, 95%CI 1.2-3.7), missed work days were 2.3 times higher, and cathelicidin concentrations were significantly lower (OR 0.49) (PMID 39301110). The apparent contradiction reflects two sides of the same story: short-term supplementation cannot reverse an already established baseline state, but the baseline state itself is the strongest risk marker. A 2021 summary in The Lancet Diabetes & Endocrinology aggregating 46 RCTs and 75,541 people corroborated this—overall protective effect 8% (OR 0.92), with the strongest subgroup protection observed in daily low-dose regimens (400-1000 IU) (OR 0.70) (PMID 33798465).
The observational data similarly points to “baseline stratification” rather than “equal treatment.” The 1958 British birth cohort of 6,789 individuals demonstrated strong seasonality in respiratory infections, strictly inversely correlated with seasonal serum 25(OH)D curves: a 10 nmol/L increase in 25(OH)D reduced infection risk by 7% (PMID 21736791). An Austrian study of 655 critically ill patients employed a more nuanced approach—stratifying by month-specific 25(OH)D tertiles (treating seasonal fluctuation itself as the baseline)—finding that the low tertile group had a hospital mortality rate 2.05 times higher than the high tertile group (PMID 24661739). These studies indicate that so-called “seasonality” is essentially the叠加 (superposition) of “people lacking in winter exposed to winter viruses”; seasonality is not the exposure itself but an upstream driver of exposure.
This hypothesis is not new—it originates from Hope-Simpson’s “seasonal stimulus” hypothesis proposed in 1981. After studying influenza pandemics for years, this epidemiologist posited that a “seasonal stimulus” closely related to solar radiation explains the seasonal outbreaks of influenza. A 2006 review consolidated this line: solar radiation drives cutaneous synthesis of Vitamin D, which is universally lacking in winter, while 1,25(OH)2D as a steroid hormone deeply regulates immunity—stimulating antimicrobial peptide expression, suppressing excessive inflammatory factors, and enhancing macrophage oxidative bursts; volunteers vaccinated with attenuated influenza virus in winter were more prone to fever and serological reactions; Vitamin D supplementation reduced respiratory infections in children (PMID 16959053). Note the structure of this hypothesis: it does not claim “solar radiation directly kills viruses,” but rather “solar radiation alters host defenses via Vitamin D”—in other words, the host-end explanation for influenza seasonality is inherently the “deficient suffer from the pathogen” model. The concept of “time-energy causing disease” in the Five Movements and Six Qi may correspond to this at the molecular level.
This directly connects to the Five Movements and Six Qi concept of “the deficient suffer from the pathogen.” Traditional Chinese medicine states “Wherever the pathogen gathers, there must be deficiency of vital energy”—for the same temporal qi (solar cycle, climatic rhythm), those with sufficient vital energy experience “the four seasons’ qi each transforms according to its nature,” while those with deficient vital energy cannot avoid “the deficient evil and thief winds, which must be avoided at certain times.” Translated into modern language: the effect magnitude of environmental exposure is not constant but a function of baseline status. On the chain Solar Cycle → UV → Vitamin D, “deficient are sensitive, sufficient are immune” has been repeatedly confirmed at the RCT level (PMID 28202713/39301110); thus, a natural inference is that the impact of the solar activity year (low UVB flux, longer winter) on health outcomes should concentrate within populations with baseline 25(OH)D deficiency—a hypothesis directly testable in existing cohorts.
Why is this hypothesis worth a dedicated research topic? Because its statistical significance is obscured by mediation analysis. Mediation analysis answers “what proportion of the effect is mediated via Vitamin D,” implicitly assuming homogeneity across the population; effect modification answers “whether this mediation holds only for a subset,” allowing for population heterogeneity. Both can coexist: the overall mediation proportion may be low (due to non-Vitamin D pathways, see uv-pathway-vitamin-d-dependent-independent), but the mediation proportion within the deficient subgroup may be high—this is precisely one reason why the “Solar Cycle-Disease” association has been inconsistently detected over decades: the proportion of deficient individuals varies across studies, altering the effect size. Explicitly modeling this interaction term may be the key to unlocking the heterogeneity in the Solar-Health association.
Evidence and Limitations
Verified: Baseline 25(OH)D <25 nmol/L supplementation significantly reduces acute respiratory infection (OR 0.58), while ≥25 nmol/L shows no benefit (PMID 28202713); baseline insufficiency (<50 nmol/L) confers 2.1 times higher infection risk (OR 2.1), lower cathelicidin levels, and short-term supplementation fails to reverse this (PMID 39301110); linear negative correlation between 25(OH)D and infection risk, inverse seasonal curves (PMID 21736791); low tertiles stratified by season-specific thirds show critically ill mortality HR 2.05 (PMID 24661739); overall supplement OR 0.92, daily regimen subgroup OR 0.70 (PMID 33798465); “seasonal stimulus” framework holds, winter supplementation reduces pediatric respiratory infections (PMID 16959053).
Partially Verified: The “Solar Cycle-Disease” association exists but effect sizes are unstable (see sunspot-latitude-health-effect-comparison); “proportion of deficient individuals causing effect size drift” is a candidate mechanism for explaining heterogeneity; winter/high-latitude deficiency risk determined jointly by latitude, skin pigmentation, and GC genotype (rs4588) (PMID 33230496), providing a heritable stratification variable for “who is more susceptible to solar cycle effects.”
To Be Verified: Explicit interaction modeling of Solar Cycle Phase (sunspot number/UVB flux) × Baseline 25(OH)D (whether the “Solar-Disease” association is significantly stronger in the deficient subgroup); joint decomposition of effect modification and mediation proportion (whether mediation share is higher in deficient subgroups than the total population); cross-latitude gradient (high-latitude high deficiency → stronger solar effects); Mendelian randomization design using GC rs4588 genotype as an instrumental variable for “baseline status.”
Related Reading
- How Large Is the Mediating Role of Serum Vitamin D in the Solar Cycle’s Impact on Health?
- Splitting the Contribution of UV Pathways: Vitamin D-Dependent vs. Non-Dependent in Solar Cycle Effects
- Cross-Latitude Multi-Center Direct Comparison of Solar Activity Effects
- Five Movements and Six Qi = Climate Cycle Medicine
- Research Pool: Environment-Life Frequency Interface
Evidence Level: A (Effect modification itself has direct PubMed support: IPD meta-subgroup analysis PMID 28202713, RCT baseline stratification PMID 39301110; “Solar Cycle × Baseline Status Interaction Term” lacks direct testing, representing an extended inference)
References: PMID 28202713, PMID 39301110, PMID 33798465, PMID 21736791, PMID 24661739, PMID 16959053, PMID 33230496
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.
Leave a Reply