Food · Misleading · 营销标签
Plant Milk
燕麦奶杏仁奶豆奶不是同一个东西 · 蛋白量质差一个数量级 · 钙和 B12 只有强化款才有 · 植物奶≠更健康, 标签说了算
Last updated
Story path
- 1What plant milk is · one word, four very different drinksWhat plant milk is · one word, four very different drinks
- 2Macros · protein differs by an order of magnitudeMacros · protein differs by an order of magnitude
- 3Fat · oat and almond low; coconut flips itFat · oat and almond low; coconut flips it
- 4Rich in what · calcium and B12 only when fortifiedRich in what · calcium and B12 only when fortified
- 5What it lacks · how to pairWhat it lacks · how to pair
- 6Key knowledge · how 'healthier' is packagedKey knowledge · how 'healthier' is packaged
- 7How to choose · the three label linesHow to choose · the three label lines
- 8Debunking · three common pitchesDebunking · three common pitches
Chapter 1
What plant milk is · one word, four very different drinks
What plant milk is · one word, four very different drinks
Plant-based milk is a beverage category, not a single food: plant material ground, filtered and formulated with water. The four common types have completely different nutritional structures:
Soy milk: high protein, close to cow's milkOat milk: high carbohydrate, low protein, naturally sweetAlmond milk: extremely low-calorie, extremely low protein, mostly waterCoconut drink: saturated fat, essentially no protein
Why it is booming: lactose intolerance (see milk's mechanism), sustainability narratives, veganism, and the intuition that 'plant = healthier'. The first two are legitimate motives; the last is exactly what this story takes apart.
The core position: plant milk is not a natural substitute for milk; it is a set of different products. To judge them, ignore the word 'plant' and read the ingredient list and nutrition panel — that is what this story teaches.
Soy milk: high protein, close to cow's milkOat milk: high carbohydrate, low protein, naturally sweetAlmond milk: extremely low-calorie, extremely low protein, mostly waterCoconut drink: saturated fat, essentially no protein
Why it is booming: lactose intolerance (see milk's mechanism), sustainability narratives, veganism, and the intuition that 'plant = healthier'. The first two are legitimate motives; the last is exactly what this story takes apart.
The core position: plant milk is not a natural substitute for milk; it is a set of different products. To judge them, ignore the word 'plant' and read the ingredient list and nutrition panel — that is what this story teaches.
机制 · 一条流水线决定了全部差别
四种植物奶的营养差得像四种东西, 但它们走的是同一条流水线。看懂这条线, 后面每一幕的数字你都能自己解释。泡: 原料先泡水泡软, 让细胞壁松开, 后面才磨得动。
磨: 加水磨成浆。这一步最关键的其实是加了多少水 —— 原料占比越低, 后面每一行营养都被按同一个比例稀释。
酶: 只有燕麦这类淀粉多的原料需要。加一种专门剪淀粉的酶, 把糊状的淀粉剪短, 浆才从稠变稀。剪出来的短糖是甜的 —— 燕麦奶不加糖也带甜味, 就是在这一步做出来的。
滤: 用滤网把渣挡下来。渣里主要是纤维, 也带走一部分蛋白。留在杯子里的, 只是穿得过滤网的那部分。
调: 加油调口感、加盐调风味、加胶体调稠度、加钙和维生素做强化。这一步是配方而不是原料 —— 所以同名的两盒可以差得很远。
均质与灭菌: 把油滴打碎、加热杀菌, 让它常温也能放几个月。
一句话: 配料表就是这条流水线的说明书。你在货架前读的每一行, 都对应上面某一道工序里有人做过的一个决定。
误区 · 一长串配料不等于不健康
磨完滤完的浆有一个纯物理的问题: 油比水轻, 会浮上去; 固体颗粒比水重, 会沉下去。放两天, 一盒植物奶就分成上下两层。厂里用两个办法按住它。一个是均质: 让液体高压穿过一道极窄的缝, 把油滴撕成极小的颗粒 —— 颗粒越小, 浮得越慢。另一个是胶体: 加一点吸水后能把水变稠的植物胶, 水一稠, 颗粒就浮不动也沉不动。乳化剂做的是第三件事: 它一头亲油、一头亲水, 停在油滴表面, 让油滴之间不容易再抱成大滴。
看清这一层之后, 配料表就好读了: 胶体、乳化剂、酸度调节剂这些名字长的东西, 干的是物理稳定的活, 和这盒东西有没有营养基本无关。它们既不加分也不减分。
真正决定营养的仍然是三行数字: 钙、蛋白、糖。读配料表是为了知道厂家做了什么, 读营养成分表才是为了知道你喝到了什么。
Chapter 2
Macros · protein differs by an order of magnitude
Macros · protein differs by an order of magnitude
Protein first — the largest gap between plant milks (per 100 ml, approximate):
Cow's milk: ~3.3 gSoy milk (unsweetened): ~3.0 gOat milk (unsweetened): ~0.5-1.0 gAlmond milk (unsweetened): ~0.4-0.6 g
So a 250 ml glass of oat or almond milk has one-sixth to one-tenth the protein of milk. For anyone using milk as a daily protein source, switching to oat or almond milk is a big protein cut.
Calories and carbohydrate flip the other way: unsweetened oat milk runs ~5 g carbs/100 ml (USDA measured), sweetened higher — oats bring starch, and the enzymatic step in production cuts that starch into shorter sugars, which is why oat milk tastes sweet without added sugar; almond milk is very low-calorie (~15 kcal/100 ml), the result of 'almost nothing added'.
Soy milk is the only one close to milk on macros; oat milk is a 'carb drink', almond milk is 'flavored water'. Remember these three sentences, then look at the rest.
Cow's milk: ~3.3 gSoy milk (unsweetened): ~3.0 gOat milk (unsweetened): ~0.5-1.0 gAlmond milk (unsweetened): ~0.4-0.6 g
So a 250 ml glass of oat or almond milk has one-sixth to one-tenth the protein of milk. For anyone using milk as a daily protein source, switching to oat or almond milk is a big protein cut.
Calories and carbohydrate flip the other way: unsweetened oat milk runs ~5 g carbs/100 ml (USDA measured), sweetened higher — oats bring starch, and the enzymatic step in production cuts that starch into shorter sugars, which is why oat milk tastes sweet without added sugar; almond milk is very low-calorie (~15 kcal/100 ml), the result of 'almost nothing added'.
Soy milk is the only one close to milk on macros; oat milk is a 'carb drink', almond milk is 'flavored water'. Remember these three sentences, then look at the rest.
机制 · 燕麦奶的甜, 是剪出来的
燕麦奶不加糖也甜, 很多人以为是厂家偷偷加了糖。不是。那点甜是在工厂里从燕麦自己身上剪下来的。燕麦的碳水主体是淀粉, 而淀粉是一长串葡萄糖手拉手连成的链子。整条链子放在嘴里不甜 —— 舌头上尝甜的那些接收器只认小块的糖, 认不出长链, 所以你嚼生米、嚼面粉都不觉得甜。
做燕麦奶时会加一种专门剪淀粉的酶 (淀粉酶)。它停在链子上, 一刀一刀把长链剪成短段, 最后剪出大量麦芽糖 —— 两个葡萄糖连在一起的小块糖。小块糖舌头认得, 于是浆变甜了; 同时长链被剪断, 浆也从糊状变稀, 倒进杯子里才像奶而不像粥。甜和稀, 是同一把剪刀带来的两个结果。
这件事的两层后果值得分开看:
标签层面: 这些糖没有人倒进去, 所以无添加糖在字面上是真的; 但营养成分表的碳水那一行照样把它算进去。世卫定义的游离糖管的是厂家、厨师或你自己加进去的糖, 加上蜂蜜、糖浆和果汁浓缩液里的糖 —— 酶剪出来的这一份, 正好卡在定义的缝里。你读到的不是一个骗局, 是一个定义没盖住的角落。身体层面: 把长链剪短这道工序, 本来该由你的唾液和小肠来做。燕麦奶等于把它提前到工厂里做完了 —— 液体到你嘴里的时候, 拆解的活已经替你干掉了一大半。
明白这一步, 你还能自己往下推两件事。一是为什么咖啡师版往往更稠更甜: 它要能打出奶泡、要能扛住咖啡, 靠的是加油、加胶体和调甜度, 这些都写在配料表上。二是为什么燕麦奶不像燕麦粥那样顶饱: 粥里的淀粉还是长链, 燕麦的纤维也还在; 而燕麦奶的链已经被剪短, 纤维大半留在了滤网上。
机制 · 杏仁奶的蛋白去哪了
杏仁本身不是低蛋白的食物, 它是一颗把能量存成油的种子, 蛋白也不算少。可到了纸盒里, 蛋白那一行几乎见底。中间发生了两件事。第一件是稀释。 一盒杏仁奶里, 杏仁只占很小一部分, 剩下几乎全是水。原料占多少, 就是这盒营养的天花板 —— 有些牌子会把这个比例印在包装上, 那往往是整张标签上最诚实的一处。
第二件是过滤。 磨出来的浆要用滤网把渣挡下来。渣里是纤维, 同时也带走一部分蛋白。穿得过滤网、留在杯子里的, 是被水稀释过、又被筛过一道的杏仁。
所以热量低不是杏仁的功劳, 是水的功劳。反过来说也成立: 你买的这一杯里, 杏仁放得越少, 它越便宜, 也越接近调味水。
对照豆奶就能看清同一套工序为什么结果不同。大豆是一颗把能量存成蛋白的种子, 那些储备蛋白遇水会散开、跟着浆一起穿过滤网; 杏仁的能量主要锁在油里, 磨出来的浆带走的是油和水。同样的泡、磨、滤三步, 豆浆带得走蛋白, 杏仁浆带不走。这就是四种植物奶里只有豆奶在蛋白上接近牛奶的原因 —— 不是它的配方更用心, 是原料本来就不一样。
Chapter 3
Fat · oat and almond low; coconut flips it
Fat · oat and almond low; coconut flips it
Fat is the easiest metric to misread in plant milk:
Oat milk: not high but not nothing (~2.75 g/100 ml in USDA's unsweetened refrigerated record), and some brands add rapeseed/sunflower oil for mouthfeel, pushing it higher — read the ingredient listAlmond milk: low fat, but the low calories come from using little almond, not from almonds being low-fatCoconut drink: high saturated fat (~2-3 g/100 ml), the same story as coconut oil (see coconut-oil)Soy milk: moderate fat, mostly unsaturated
For most people, plant-milk fat is not the main issue; the real reads are added sugar and protein.
'Zero cholesterol' is a common plant-milk pitch — essentially filler: plant drinks are naturally cholesterol-free, but milk's cholesterol barely moves blood lipids in most people (aha-2020-dietary-cholesterol). Selling that point is making a non-issue the headline.
For fat structure, dive to fats-omega-3 and omega-3-6-balance.
Oat milk: not high but not nothing (~2.75 g/100 ml in USDA's unsweetened refrigerated record), and some brands add rapeseed/sunflower oil for mouthfeel, pushing it higher — read the ingredient listAlmond milk: low fat, but the low calories come from using little almond, not from almonds being low-fatCoconut drink: high saturated fat (~2-3 g/100 ml), the same story as coconut oil (see coconut-oil)Soy milk: moderate fat, mostly unsaturated
For most people, plant-milk fat is not the main issue; the real reads are added sugar and protein.
'Zero cholesterol' is a common plant-milk pitch — essentially filler: plant drinks are naturally cholesterol-free, but milk's cholesterol barely moves blood lipids in most people (aha-2020-dietary-cholesterol). Selling that point is making a non-issue the headline.
For fat structure, dive to fats-omega-3 and omega-3-6-balance.
机制 · 燕麦奶为什么要加油
配料表上那行菜籽油或葵花籽油, 不是为了营养, 是为了口感。脂肪在嘴里干的是一件很物理的事: 它裹住舌面, 让液体滑过去, 而不是冲过去。牛奶之所以喝起来是顺的, 一半功劳在它的乳脂小球上。滤过之后的燕麦浆几乎没有油, 直接喝像淡淡的米汤; 加一点植物油, 再用均质机把油撕成小滴均匀悬起来, 口感立刻就往奶那边靠。
所以读到这行油, 该读出三件事: 这盒的脂肪和热量会被抬高一点; 抬上去的是不饱和脂肪, 不是饱和脂肪; 以及它是质地工程, 不是营养承诺。
同样的道理反过来也能用: 一盒宣称零脂肪的植物奶, 大概率喝起来偏水。这不是缺点, 是取舍 —— 你只需要先知道自己想要哪一头。
误区 · 零胆固醇为什么是废话
零胆固醇是植物奶包装上的常客。它是真的, 也是没用的 —— 因为它回答的不是那个要紧的问题。血里的胆固醇, 大部分不是你吃进去的, 是你的肝脏自己造的。真正把坏胆固醇 (LDL) 顶高的是饱和脂肪和反式脂肪: 它们会让肝细胞表面那些把坏胆固醇从血里捞回来的接口变少。接口一少, 血里的坏胆固醇被捞走的速度就慢下来, 于是它在血管里待得更久、堆得更高。食物里的胆固醇本身也有影响, 但走的是另一条弱得多的路。
明白了这一条, 你就能自己判断一杯椰饮: 它天然零胆固醇, 但里面装的是饱和脂肪 —— 按上面的机制, 对血脂它未必比牛奶友好。零胆固醇回答的是那个不重要的问题, 而它旁边那一行饱和脂肪, 才是重要的那个。
同一把尺子还能量很多东西: 薯片零胆固醇, 椰子油也零胆固醇。凡是植物做的, 这一行天生就是零 —— 一个所有同类都满足的条件, 不构成任何一家的优点。
Chapter 4
Rich in what · calcium and B12 only when fortified
Rich in what · calcium and B12 only when fortified
The scene where plant milk most easily deceives: calcium and vitamin B12 are added later, not native to plants.
Calcium: milk naturally has ~120 mg/100 ml. Plant milk reaches that level only when fortified — the records USDA holds are all fortified commercial products and measure 148-184 mg/100 ml, a figure that comes from the fortificant rather than from almonds or oats; unfortified versions are low. Calcium mechanisms live in calcium and bone.Vitamin D: likewise only when fortified; milk is D-fortified in most countries — compare labels.Vitamin B12: plant drinks are naturally near-zero; only fortified versions have it (see nori for the honest plant-B12 question). Vegans relying on unfortified plant milk for B12 will come up short.
Soy milk has one genuine native advantage: soy protein is a rare complete plant protein, PDCAAS ~0.91 vs milk's ~1.0 (ginting-2011-soy-pdcaas). Oat and almond protein is low in both amount and quality.
Three label reads: ① check 'fortified' and the actual calcium figure ② check protein grams ③ check added sugar. Three reads and the plant milk reveals itself.
Calcium: milk naturally has ~120 mg/100 ml. Plant milk reaches that level only when fortified — the records USDA holds are all fortified commercial products and measure 148-184 mg/100 ml, a figure that comes from the fortificant rather than from almonds or oats; unfortified versions are low. Calcium mechanisms live in calcium and bone.Vitamin D: likewise only when fortified; milk is D-fortified in most countries — compare labels.Vitamin B12: plant drinks are naturally near-zero; only fortified versions have it (see nori for the honest plant-B12 question). Vegans relying on unfortified plant milk for B12 will come up short.
Soy milk has one genuine native advantage: soy protein is a rare complete plant protein, PDCAAS ~0.91 vs milk's ~1.0 (ginting-2011-soy-pdcaas). Oat and almond protein is low in both amount and quality.
Three label reads: ① check 'fortified' and the actual calcium figure ② check protein grams ③ check added sugar. Three reads and the plant milk reveals itself.
数字 · USDA 记录里的那个钙, 是谁的
USDA 食品成分库收录的植物奶, 全都是强化过的市售商品。测出来的钙在 148-184 mg/100 ml, 比牛奶还高一截。但这个数字不是杏仁的, 也不是燕麦的 —— 它来自厂家往液体里加进去的钙。同一颗杏仁, 如果没人加钙, 这一行就会掉到很低。
这件事值得单独记住, 因为它是营养成分表的一个通用性质: 表测的是这盒液体现在的样子, 不是原料的样子。 于是读标签有了一个稳定的分类法 —— 钙和维生素 D 这类强化项在植物奶里几乎全靠加, 而蛋白那一行反倒骗不了人: 除非包装明写了蛋白强化, 没有厂家会往里补蛋白, 它老老实实反映原料放了多少。
读标签的三步, 按这个顺序走: ① 看有没有强化字样和具体钙含量 ② 看蛋白多少克 ③ 看添加糖。三步走完, 一盒植物奶是什么, 立刻现形。
机制 · 为什么植物天生没有这一种
钙和维生素 B12 缺的方式不一样: 钙是植物含量太少, 而这一种是植物根本不做。能从零把它造出来的只有细菌。牛羊的胃里养着一整套微生物, 微生物造出来的这一份被吸收进牛羊的肉和奶, 人再从肉和奶里拿到 —— 你身上的每一分, 追到源头都是某个细菌的产物。
一株燕麦、一颗杏仁、一粒大豆, 身上没有这套微生物, 自己也不需要它。所以植物性饮品这一行天然就是零: 不是加工过程弄丢的, 是从来就没有过。 强化款里补进去的那一份, 同样来自细菌发酵 —— 只不过发酵槽在工厂里, 不在牛的胃里。
这条机制真正的用处, 是让你能自己判断别的说法: 任何声称天然富含 B12 的植物性食品, 都值得多问一句 —— 那个数字是从哪测出来的、是不是身体真能用的形式 (见 nori 那篇对植物性 B12 的诚实之问)。
机制 · 强化的钙不会自己化在水里
强化用的钙多半是不太溶于水的钙盐 —— 它并没有化开, 只是被胶体托着, 悬在液体里。放久了, 它会慢慢往盒底沉。所以有一个动作值得养成: 喝之前把盒子摇匀。 不摇的话, 前半盒偏稀、最后一口偏浓 —— 你付了强化款的钱, 却没喝到强化款的钙。这大概是这一整篇里最便宜的一个改进: 两秒钟, 不花钱。
到了小肠, 钙的来路就不重要了。无论它来自牛奶还是来自强化剂, 都要先变成钙离子, 才能被肠壁上的通道搬进血里, 身体分不出它的出身。真正影响吸收的是同一口食物里有没有东西挡着它: 草酸和植酸会在肠道里跟钙抱成一团, 一起被排掉, 那部分钙等于没吃。
再往下推一步就是: 一盒强化了钙的植物奶, 如果你拿它配一碗高草酸的菜同时下肚, 标签上那个漂亮的数字会打折。吸收多少, 从来不只取决于吃进去多少。
蛋白 · 豆奶为什么是唯一的例外
豆奶有一个天然加分项, 而且不靠强化: 大豆蛋白是植物里少见的完全蛋白, 蛋白质消化校正氨基酸评分 (PDCAAS) 约 0.91, 接近牛奶的 1.0 (ginting-2011-soy-pdcaas)。燕麦和杏仁的蛋白则是量也少、质也弱。完全蛋白是什么意思? 你的身体每造一条新蛋白, 都要按图纸把二十种氨基酸拼起来, 其中有九种身体自己造不出来, 只能从食物里拿。少了任何一种, 拼装就在那一步停住 —— 其它氨基酸再多也补不上这个缺口, 多出来的部分会被拆掉当能量烧掉。大多数植物蛋白都缺那么一两种; 大豆是少数不缺的。
PDCAAS 这个分数把两件事合在一起打: 氨基酸配得齐不齐, 加上人体能消化吸收多少。牛奶是满分参照; 大豆的分数意味着它在这两项上都紧贴着牛奶。
所以那句豆奶接近牛奶不是营销话, 它背后有一套测量方法。而同一句话换成燕麦奶或杏仁奶就不成立 —— 它们的蛋白既没有量, 也没有这个分数。
Chapter 5
What it lacks · how to pair
What it lacks · how to pair
Plant milk's gaps are structural, and the fixes differ by population:
Protein gap: oat/almond milk is too low to be a milk substitute; soy or 'protein-fortified' versions close itCalcium gap: on unfortified plant milk, 300 ml a day loses you a hundred-plus mg of calcium; choose fortified, or get it from leafy greens, tofu and sesame (kale covers the calcium-absorption comparison)B12 gap: vegans must rely on fortified versions or supplementsIodine: milk is an important iodine source in some regions; switching to plant milk may cut iodine intake — minor in coastal/salted-iodine areas, worth watching inland (see kelp on iodine)
For young children especially: unfortified oat/almond drinks are very low in protein and calcium, so using them as a full cup-for-cup swap for cow's milk can cut both. Toddler 'milk' swaps are not casual — choose soy or explicitly fortified products, and consult a pediatrician.
In one line: every plant-milk gap has a fix, but fixing it is your job, not the product's default.
Protein gap: oat/almond milk is too low to be a milk substitute; soy or 'protein-fortified' versions close itCalcium gap: on unfortified plant milk, 300 ml a day loses you a hundred-plus mg of calcium; choose fortified, or get it from leafy greens, tofu and sesame (kale covers the calcium-absorption comparison)B12 gap: vegans must rely on fortified versions or supplementsIodine: milk is an important iodine source in some regions; switching to plant milk may cut iodine intake — minor in coastal/salted-iodine areas, worth watching inland (see kelp on iodine)
For young children especially: unfortified oat/almond drinks are very low in protein and calcium, so using them as a full cup-for-cup swap for cow's milk can cut both. Toddler 'milk' swaps are not casual — choose soy or explicitly fortified products, and consult a pediatrician.
In one line: every plant-milk gap has a fix, but fixing it is your job, not the product's default.
机制 · 换掉一杯, 掉的不只是那一杯
缺口这件事最反直觉的地方在于: 它往往不是你少吃了什么, 而是你用什么占掉了那个位置。一杯饮料在一天里占掉两样东西: 胃里的空间, 和你心里那个写着这顿有奶了的勾。把牛奶换成未强化的杏仁奶, 胃一样饱, 勾一样打上, 但蛋白和钙那两行悄悄归了零。你不会感觉到缺 —— 缺口没有味道, 也不当场发作, 它要几个月甚至几年才在别处显形。
这就是为什么替代比增加更需要小心: 往饮食里加一样东西, 最坏的结果是白加; 拿一样东西去换掉另一样, 最坏的结果是把原来那一份也一起端走了。
所以给自己换一个更好的问题。不问这个东西健康吗, 而问它顶掉了什么。同一杯燕麦奶, 换掉的是一杯含糖饮料还是一杯牛奶, 是两件完全不同的事 —— 前者是升级, 后者是降级。
人群 · 为什么幼儿最经不起这一换
同一个替换, 落在成年人和幼儿身上完全不是一回事。原因有两个, 都很物理。一是需求高。幼儿的身体正在快速地长骨头、长肌肉, 按体重折算, 他们每公斤需要的蛋白和钙都比成年人多。
二是胃小。他们一天能装下的食物总量有限, 每一口的营养密度都比成年人金贵。一杯液体在成人的一天里只是配角, 在幼儿那里可能就是主食之一。
两条合起来看: 同一杯几乎不含蛋白的杏仁奶, 在成人身上是少了一点, 在幼儿身上可能是主食被换成了水。而且幼儿不会告诉你他缺蛋白, 他只会告诉你他饱了 —— 饱腹感是由体积和液体量决定的, 和营养密度没有关系。
所以这件事在幼儿身上要按医疗决定来处理, 不按口味偏好来处理。正文那句咨询儿科医生不是客套话, 它是这一整幕唯一不能自己拿主意的地方。
Chapter 6
Key knowledge · how 'healthier' is packaged
Key knowledge · how 'healthier' is packaged
The core plant-milk pitch is 'healthier than milk'; this story's key knowledge is breaking it into four testable claims:
1. Greener: largely true (plant drinks have lower carbon than milk) — an environmental claim, not a nutritional one
2. Less saturated fat/cholesterol: true but often filler (see fat-quality)
3. More nutritious: possible only for specific products (fortified + unsweetened + decent protein like soy), and never automatic
4. Better for lactose intolerance: true — plant milk is naturally lactose-free; its most genuine selling point (the mechanism lives in milk and nih-niddk-lactose-intolerance)
The real problem hides in product formulation: oat milk often gets rapeseed oil and salt for mouthfeel; sweetened versions add sugar (sucrose/maple/concentrated juice) until a glass's free sugars rival a soft drink; 'barista' editions are typically sweeter and thicker.
There is also a psychological trap: consumers read 'plant milk' as a health symbol, assuming oat milk is 'cleaner' than milk — but nutrition does not recognize symbols, only the three numbers on the label (calcium, protein, sugar).
Conclusion: plant milk is a diverse, formulation-driven category, not a synonym for the conclusion 'healthier'. The legitimate reasons to choose it (lactose intolerance, veganism, sustainability) all stand; but each must be verified on the label rather than delegating thought to the word 'plant'.
1. Greener: largely true (plant drinks have lower carbon than milk) — an environmental claim, not a nutritional one
2. Less saturated fat/cholesterol: true but often filler (see fat-quality)
3. More nutritious: possible only for specific products (fortified + unsweetened + decent protein like soy), and never automatic
4. Better for lactose intolerance: true — plant milk is naturally lactose-free; its most genuine selling point (the mechanism lives in milk and nih-niddk-lactose-intolerance)
The real problem hides in product formulation: oat milk often gets rapeseed oil and salt for mouthfeel; sweetened versions add sugar (sucrose/maple/concentrated juice) until a glass's free sugars rival a soft drink; 'barista' editions are typically sweeter and thicker.
There is also a psychological trap: consumers read 'plant milk' as a health symbol, assuming oat milk is 'cleaner' than milk — but nutrition does not recognize symbols, only the three numbers on the label (calcium, protein, sugar).
Conclusion: plant milk is a diverse, formulation-driven category, not a synonym for the conclusion 'healthier'. The legitimate reasons to choose it (lactose intolerance, veganism, sustainability) all stand; but each must be verified on the label rather than delegating thought to the word 'plant'.
配方 · 同一个名字, 两盒可以差得像两种饮料
更有营养那一条之所以不天然成立, 是因为植物奶真正的变量不在原料, 在配方。同一个燕麦奶的名字底下, 两盒的差别可以比燕麦奶和牛奶的差别还大。常见的三处配方动作:
加油: 燕麦奶为了口感常加菜籽油或葵花籽油, 脂肪和热量跟着上去 —— 那是质地工程, 不是营养升级加盐: 盐在这里不是为了咸。少量的盐会让甜味和香味更突出, 是食品配方里很老的一个手法加甜: 甜味款加的糖可以是蔗糖、枫糖浆或浓缩果汁。加到一定量, 一杯的游离糖能赶上一杯含糖饮料; 咖啡师版为了打奶泡和配咖啡, 往往更甜也更稠
第三条有个容易漏掉的坑: 世卫定义的游离糖不只是白砂糖, 还包括蜂蜜、糖浆, 以及果汁浓缩液。所以一盒用浓缩果汁调甜的植物奶, 配料表上可以完全看不到糖这个字, 按世卫的定义它照样是游离糖。
想绕过这个坑, 办法很简单: 别在配料表里找糖字, 直接看营养成分表里那一行数字。配料表告诉你厂家用了什么, 只有营养成分表告诉你它变成了多少。
误区 · 一个词为什么能关掉判断
这一幕最值钱的不是哪个牌子好, 而是看清一个思维捷径。它是这样工作的: 一个食物一旦被贴上某个正面标签, 大脑会顺手把这个标签延伸到它身上所有别的属性 —— 看到植物基三个字, 就默认它同时低糖、低脂、营养高。这三件事之间其实没有任何因果关系。植物油是植物的, 白砂糖也是植物的。
标签贴上之后还有第二层影响: 分量容易失控。反正它是健康的这句话, 是很多人多倒一杯的理由; 而一杯甜味燕麦奶天天两三杯地喝, 累计下来的糖未必比偶尔一次含糖饮料少。
破除它的办法不是背一张好牌子清单 —— 清单会过期, 配方会改。办法是换一个提问方式: 不问这是不是健康食品, 而问这一盒的钙、蛋白、糖各是多少。前一个问题没有答案, 后一个问题就印在包装背面, 几秒钟就能读完。
结论 · 三个正当理由, 三种验证动作
把这一幕收成一句能带走的话: 植物奶是一个类别多样、由配方决定的饮品家族, 不是更健康这个结论的代名词。选它不选牛奶的三个正当理由都成立 —— 乳糖不耐、素食、环保。但每一个理由对应的验证动作不一样:
因为乳糖不耐而选: 这条几乎不用验证, 四类植物奶天然都不含乳糖因为素食而选: 必须验证强化, 尤其是那两项植物天生给不了的 (钙靠强化补足, 另一项只有细菌造得出来)因为环保而选: 环境账和营养账是两本账。环保成立不代表这盒的营养也成立 —— 营养仍然要回到那三行数字
换句话说, 正当理由决定要不要换, 标签决定换成哪一盒。这两个决定分开做, 植物奶这件事就不再让人困惑了。
Chapter 7
How to choose · the three label lines
How to choose · the three label lines
Turning judgment into an executable selection flow:
Define the use first: lactose-intolerant seeking a substitute (soy or oat both work) · vegan needing nutrition (soy + fortified) · coffee companion (mouthfeel first) · cutting calories (almond, but expect no nutrition).
Then read the three lines:
1. Calcium: ≥100 mg/100 ml counts as 'usable like milk'
2. Protein: ≥3 g/100 ml approaches milk (soy usually passes; oat and almond usually do not)
3. Sugar: unsweetened or low; treat sweetened versions as soft drinks
Scan the ingredients: water + plant material + fortifiers is the clearest structure; a long string of emulsifiers, gums, flavors and oils is not more nutritious — nutrition still lives in the calcium/protein/sugar lines.
Population notes: for toddlers choose soy/fortified and ask a pediatrician; thyroid patients mind the soy-iodine interaction (see soy-tofu's timing advice); people on medication should separate calcium-rich plant milk from drug intake, as calcium can interfere with absorption.
In one line: delete the words 'plant milk' from your cart, add the three numbers (calcium, protein, sugar), and the choice becomes clear.
Define the use first: lactose-intolerant seeking a substitute (soy or oat both work) · vegan needing nutrition (soy + fortified) · coffee companion (mouthfeel first) · cutting calories (almond, but expect no nutrition).
Then read the three lines:
1. Calcium: ≥100 mg/100 ml counts as 'usable like milk'
2. Protein: ≥3 g/100 ml approaches milk (soy usually passes; oat and almond usually do not)
3. Sugar: unsweetened or low; treat sweetened versions as soft drinks
Scan the ingredients: water + plant material + fortifiers is the clearest structure; a long string of emulsifiers, gums, flavors and oils is not more nutritious — nutrition still lives in the calcium/protein/sugar lines.
Population notes: for toddlers choose soy/fortified and ask a pediatrician; thyroid patients mind the soy-iodine interaction (see soy-tofu's timing advice); people on medication should separate calcium-rich plant milk from drug intake, as calcium can interfere with absorption.
In one line: delete the words 'plant milk' from your cart, add the three numbers (calcium, protein, sugar), and the choice becomes clear.
机制 · 为什么是这三行, 不是别的行
三条线不是随手画的, 每条背后都有一个身体上的理由。钙那条线画在牛奶的水平上, 因为这杯东西是去顶替牛奶在你饮食里原来那个位置的。低于这个量级, 它可以是一杯好喝的饮料, 但它接不住原来那一份钙 —— 位置被占了, 功能没接上, 这是最隐蔽的一种亏。
蛋白那条线同理: 蛋白是身体每天都在拆和造的东西, 它没有像脂肪那样的现成仓库。今天少了就是少了, 明天多吃也补不回昨天没造的那一份。所以蛋白这一行不该交给一杯饮料去碰运气。
糖那条线要单独说, 因为液体有个特殊性质: 它下去得快, 也不太占胃。同样多的糖装在固体里, 你会先觉得饱、会自动停下; 装在液体里, 停的信号来得晚。所以同一克糖, 在杯子里比在盘子里更容易被喝多。
这三条线还有一个好处是互相独立: 一盒可以钙达标而蛋白不达标 (强化的燕麦奶就是), 也可以蛋白达标而糖超标 (甜味豆奶就是)。三行一起读, 才拼得出这盒东西的全貌; 只看一行, 每次都会被那一行骗一次。
厨房 · 豆奶倒进咖啡为什么会絮
这一条既是使用技巧, 也是一次免费的蛋白检测。咖啡是酸的。豆奶里的大豆蛋白平时靠表面带的同种电荷互相推开, 才能均匀散在水里; 一遇到酸, 这些电荷被中和掉, 蛋白之间不再互斥, 就会抱成一团一团的絮 —— 看着像坏了, 其实只是蛋白被析了出来。热还会加速这件事, 所以热咖啡比冷咖啡更容易絮。
能絮, 恰恰说明里面有蛋白。 杏仁奶倒进咖啡几乎不絮, 不是因为它更稳定, 是因为里面几乎没有蛋白可絮。下次看到一盒号称绝不分层的植物奶, 可以顺手想一想: 它凭什么这么稳。
想避免絮又不想换奶, 有三个纯物理的办法: 让豆奶先回到室温再用 (温差越小, 蛋白越不容易一下子析出); 把咖啡慢慢倒进奶里, 而不是把奶倒进一整杯咖啡 (前者让酸一点点被稀释, 后者是让第一口奶直接撞上满杯的酸); 或者选咖啡师版 —— 它们常加了酸度调节剂, 干的正是把这一步按住的活。
Chapter 8
Debunking · three common pitches
Debunking · three common pitches
Three pitches taken apart:
Pitch one: oat milk provides calcium like milk. True only for fortified versions, and only on the numbers. Unfortified oat milk has near-zero calcium; fortified reaches 120 mg and above (USDA's fortified commercial records measure 148-184 mg/100 ml). 'Oat milk gives you calcium' is marketing presenting fortification as native — same logic as reading sugar on 'unsweetened' yogurt.
Pitch two: plant milk is higher-protein / cleaner. Backwards. Plant milk is generally lower-protein (oat and almond by an order of magnitude); only soy approaches milk. 'Clean' is a marketing word with no nutritional definition — read the oils and sugar in the ingredients.
Pitch three: plant milk is healthier for children. The most dangerous one. Unfortified oat/almond milk drops toddlers' protein and calcium simultaneously. Switching a toddler's milk is a decision for a clinician, not for a pitch.
The universal test for such claims: for every nutritional conclusion, ask 'which product, and which label line, supports it'. Fortified, unsweetened, soy — three words separate the pitch from the facts.
Pitch one: oat milk provides calcium like milk. True only for fortified versions, and only on the numbers. Unfortified oat milk has near-zero calcium; fortified reaches 120 mg and above (USDA's fortified commercial records measure 148-184 mg/100 ml). 'Oat milk gives you calcium' is marketing presenting fortification as native — same logic as reading sugar on 'unsweetened' yogurt.
Pitch two: plant milk is higher-protein / cleaner. Backwards. Plant milk is generally lower-protein (oat and almond by an order of magnitude); only soy approaches milk. 'Clean' is a marketing word with no nutritional definition — read the oils and sugar in the ingredients.
Pitch three: plant milk is healthier for children. The most dangerous one. Unfortified oat/almond milk drops toddlers' protein and calcium simultaneously. Switching a toddler's milk is a decision for a clinician, not for a pitch.
The universal test for such claims: for every nutritional conclusion, ask 'which product, and which label line, supports it'. Fortified, unsweetened, soy — three words separate the pitch from the facts.
第四个话术 · 燕麦奶算全谷物
这一条没进上面的三条, 但它出现得越来越多: 燕麦是全谷物, 所以燕麦奶也算全谷物。全谷物之所以被推荐, 靠的是完整这两个字 —— 麸皮、胚芽、纤维和还没被拆开的淀粉一起进肚子。消化慢, 纤维还能在肠道里带走一部分东西。
而做成燕麦奶的过程, 正好把这三样一件件拿掉了: 酶把淀粉剪短了 (不再是没被拆开的), 滤网把麸皮和大部分纤维挡在外面 (不再是完整的), 剩下穿过滤网进到杯子里的, 主要是水、短糖和一点蛋白。
所以杯子里那一杯和碗里那一碗燕麦粥, 共用的只是名字里的燕麦两个字。这不是说燕麦奶不能喝 —— 它是一杯合格的饮料; 只是它拿不到全谷物那一份好处, 而那份好处正是它被推荐时借用的理由。借来的理由, 经不起追问原料去哪了。
工具 · 三个问题, 拆掉大部分食品话术
这一篇真正想留给你的不是植物奶的结论, 是三个可以带去货架任何一处的问题。问题一: 这个营养素是长出来的, 还是加进去的? 加进去的可以随时被拿掉 —— 换一个配方、换一个批次, 它就没了。长出来的跟着原料走, 稳定得多。这个问题一问, 强化款和未强化款立刻分成两堆。
问题二: 这一行数字, 是在哪个产品上测出来的? 燕麦奶含钙这句话在强化款上是真的, 在未强化款上是假的。一句话在一部分产品上成立, 不等于它在整个类别上成立 —— 而广告说的永远是类别。
问题三: 把这句宣称删掉, 结论还站得住吗? 零胆固醇这四个字删掉之后, 那盒东西的健康程度一点没变 —— 说明它本来就没在承重。凡是删掉也不影响结论的宣称, 都是装饰。
这三个问题不需要你记住任何数字, 也不需要你相信任何人。它们唯一的要求, 是你肯把包装翻到背面。
References · 8
- U.S. Department of Agriculture, Agricultural Research Service. (2022). FoodData Central: Oat milk, unsweetened, plain, refrigerated (Foundation, FDC ID 2257046). Per 100 g: 0.8 g protein, 2.75 g fat, 5.1 g carbohydrate, 148 mg calcium, 0.51 µg vitamin B12, 1.7 µg vitamin D. The Foundation record publishes no energy value; calcium, B12 and vitamin D here are fortification, not native to oats. fdc.nal.usda.gov
- U.S. Department of Agriculture, Agricultural Research Service. FoodData Central: almond beverage, unsweetened — shelf stable (SR Legacy, FDC ID 174832) and refrigerated (Foundation, FDC ID 2257045). Per 100 g: 15 kcal, 0.4-0.66 g protein, 0.96-1.56 g fat, 0.67-1.31 g carbohydrate. Both records are fortified commercial products and report 158-184 mg calcium and 1.0-1.6 µg vitamin D per 100 g — figures that come from fortification, not from almonds. fdc.nal.usda.gov
- Hughes, G. J., Ryan, D. J., Mukherjea, R., & Schasteen, C. S. (2011). Protein digestibility-corrected amino acid scores (PDCAAS) for soy protein isolates and concentrate: criteria for evaluation. Journal of Agricultural and Food Chemistry, 59(23), 12707-12712. Soy protein is a complete protein with all nine essential amino acids and a PDCAAS near 1.0, comparable to milk and egg. 10.1021/jf203220v
- Carson, J. A. S., Lichtenstein, A. H., Anderson, C. A. M., Appel, L. J., Kris-Etherton, P. M., Meyer, K. A., et al. (2020). Dietary cholesterol and cardiovascular risk: A science advisory from the American Heart Association. Circulation, 141(3), e39-e53. Saturated and trans fats raise LDL more than dietary cholesterol itself. 10.1161/CIR.0000000000000743
- National Institutes of Health, Office of Dietary Supplements. (2024). Calcium — Fact Sheet for Health Professionals. ods.od.nih.gov/factsheets/Calcium-HealthProfessional
- National Institutes of Health, Office of Dietary Supplements. (2024). Vitamin B12 — Fact Sheet for Health Professionals. ods.od.nih.gov/factsheets/VitaminB12-HealthProfessional
- World Health Organization. (2015). Guideline: Sugars intake for adults and children. Geneva: WHO. Recommends reducing free sugars to <10% of total energy intake (strong recommendation), and suggests a further reduction to <5% (conditional). 'Free sugars' include monosaccharides and disaccharides added by the manufacturer, cook, or consumer, plus sugars naturally present in honey, syrups, fruit juices, and fruit juice concentrates — explicitly distinguishing them from the intrinsic sugars in whole fruit, vegetables, and milk. www.who.int/publications/i/item/9789241549028
- National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). (2018). Lactose Intolerance: Definition & Facts. U.S. National Institutes of Health. www.niddk.nih.gov/health-information/digestive-diseases/lactose-intolerance