How will I know if collagen is working?

The knowing was decided before the first capsule: how the trials knew, and why a reader without a baseline always gets a yes

How will I know if collagen is working is a question that is nearly always asked a few weeks after starting, and the honest answer has to begin by saying that the knowing was decided before the first capsule, because the method by which anyone has ever known whether collagen does anything is a comparison, and a comparison needs a before. The trials knew, to the extent they knew anything, because they measured before and after. A skin trial measured elasticity with a suction cutometer and hydration with a corneometer on a marked patch of forearm or cheek on day one, then again at twelve weeks, in a collagen group and a placebo group, so that the change in the collagen group could be set against the change in people who took a matching powder of nothing through the same twelve weeks of the same season; a joint trial scored pain and function on a questionnaire at baseline and at twelve and twenty-four weeks in both arms; the nail trial measured nail growth and counted broken nails before and through six months. The placebo arm is the part that readers forget and the part that matters most, because it controlled for everything else that changes over twelve weeks: the season, the sleep, the training block, the holiday, the new moisturiser, the resolution to drink more water, the simple fact of being watched, all of which moved the placebo group too, so that the collagen effect was whatever was left after the placebo group's change was subtracted, and it was, by instrument, a few percent. A reader who starts a tub in October without taking any measurement has no before, and without a before the after is compared with memory, which is the least reliable instrument in the house. And the answer that memory gives is always yes, for a reason that has nothing to do with collagen: skin, joints, nails and hair all drift. Skin looks better after a good week of sleep, a few days of drinking more water and a fortnight of not being in the wind. Joints feel better in warm weather, after a deload, after a new pair of shoes, after the stressful month ends. Nails grow faster in summer than winter and break less when the gardening stops. Hair sheds more in autumn and less in spring on a cycle that has nothing to do with anything swallowed. Every one of these improvements is real, and every one of them is credited to whatever was started most recently, which is the mechanism by which every supplement ever sold has worked for someone; the false positives are set out one by one on the first signs page, and the five signs the shelf promises are audited against the trials on the five signs page. The reader who wants to know, rather than to feel, needs a before, and the honest page supplies the three a reader can take at home.

Three baselines a reader can take at home, their trial timepoints and effect sizes, the fortnight off at the end, and the worked example

The three baselines are the home versions of the trials' instruments, and each takes less than a minute. For skin, a photograph of one patch, the back of one hand, one cheek, the crow's feet of one eye, taken in the same light, by a window at the same time of day or under the same lamp, with the same camera at the same distance and the same expression, saved with the date, because the trials measured elasticity and hydration on a marked patch and a fixed photograph is the closest a reader gets to a marked patch; two photographs twelve weeks apart in matched conditions can be compared side by side, and a reader who has them will usually see what the trials saw, which is a change too small to be sure of, and will have learned something true. For joints, a pain score out of ten on one set movement, chosen because it is the movement that hurts: the bottom of a full-depth squat, the first ten steps of the morning, the stairs at the end of the day, a kneel, scored and written down with the date, and repeated on the same movement at the same time of day, because the trials' questionnaires were exactly this, pain and function on set tasks, and a reader with a written three on the morning stairs in October can compare it with a written score in January rather than with a feeling about how the knees have been. For nails, a pen mark at the cuticle of one thumbnail, with the date, and the distance it has travelled measured a month later, since nails grow around three millimetres a month and the one nail trial measured growth and counted breaks; a second mark at the start of month four gives the comparison. Each baseline is repeated at the trial's timepoint, not before, because judging early is the second way to get the wrong answer: twelve weeks for skin, which is when the skin trials measured; twelve to twenty-four weeks for joints, which is when the joint questionnaires moved; four to six months for nails, which is how long a nail takes to grow out. And each is judged against the effect size the trials found, so that a reader knows what a win looks like and does not wait for a transformation that no trial has ever reported: a few percent in elasticity, which in a photograph is barely visible; a point or two on a ten-point pain scale, which is a four becoming a three; a millimetre or two more growth a month and fewer breaks, which is a mark travelling four millimetres instead of three. A reader whose measurements move by those amounts has seen what the trials saw and can decide whether it is worth 28p a day or a pound a day; a reader whose measurements move by nothing has also learned something, which is more than the reader without a baseline ever does. The fortnight off at the end is the last instrument: stop, repeat the three measurements two weeks later, and see whether anything slides, since the wash-out data suggest the measured effects fade over weeks after stopping, and a reader whose skin photograph, pain score and nail growth are unchanged after a fortnight off has evidence that the calendar was doing the work; the wash-out data are on the break from collagen page. The worked example is the thing to be measured against rather than trusted, stated on its true terms: two capsules, 1200mg hydrolysed type 1 marine collagen with 12mg vitamin C at 15% of its reference value plus zinc, 120 for £16.99 at 28p a day, lab tested and UK made, below the 2.5 to 10g doses at which the trials measured anything, carrying vitamin C's collagen formation sentence for normal skin function and no claim of its own, which is a reason to measure rather than a reason not to. How will I know if collagen is working: by the before you take today, the trial timepoint you wait for, the effect size you expect, and the fortnight off that tells you whose work it was.

What to check on the label