How Do I Know if Peptides Are Working? What to Measure and When

A clinical trial randomizes hundreds of people, gives half an inactive injection, and reports the difference between two averages. You cannot run that design on yourself: one person, no randomization, no blinding, no control arm. What follows recovers as much of that rigor as a single person can, and what you can conclude depends on what you wrote down before you started.

Two medications in the catalog have large placebo-controlled trials that show the size of the gap. In STEP 1, the trial of once-weekly semaglutide published in the New England Journal of Medicine in 2021, the placebo group — inactive injection plus lifestyle advice — lost 2.4 percent of body weight over 68 weeks, and 31.5 percent of those placebo participants lost at least 5 percent of their starting weight. In SURMOUNT-1, the tirzepatide obesity trial in the same journal in 2022, the placebo arm lost 3.1 percent at 72 weeks and 35 percent crossed the 5 percent mark.

Roughly a third of people who received nothing at all, in a supervised trial, hit a threshold many patients would take as proof their medication was working. That is not an argument against the medication — the semaglutide and tirzepatide groups moved far further — it is an argument that one before-and-after cannot separate a drug effect from the effect of being in a program and paying attention.

People start treatment when things are at their worst: the week the knee hurts most, the month the scale peaks, the stretch of bad sleep that prompted the consultation. Measurements taken at an extreme drift back toward that person's own average on their own, with no treatment involved. The International Journal of Epidemiology published a clear practical account of this in 2005, and the correction it recommends is the one used here: more than one baseline measurement, spaced out.

One weight is not a baseline. Body weight has a documented weekly rhythm: a 2014 analysis in Obesity Facts of nearly 4,700 daily weights from 80 adults found weight consistently higher on Sunday and Monday and falling through the working week, in people losing, maintaining and gaining alike, and concluded that weekend-to-weekday variation should be read as normal rather than as a signal.

So the baseline is seven consecutive mornings, same conditions, averaged. Do the same four weeks later.

Name one movement in writing and define it precisely enough to repeat identically: the load, the repetitions, the time of day, the warm-up. Record pain out of ten during it, range of motion measured with a phone app, and the load you managed.

For the pain number there is a published threshold. A 2001 paper in Pain, pooling 2,724 patients across ten chronic-pain trials, found that a reduction of about two points on the eleven-point scale, or about 30 percent, corresponded to patients describing themselves as much improved — a useful anchor when you are staring at a drop from 6 to 5, which on its own means very little. This baseline matters most for BPC-157, the BPC-157 + TB-500 preparation and KLOW, where published human evidence is thinnest.

For skin: three photographs, one room, one time of day, one light source, no filter, no makeup, phone at a fixed distance. Skin changes slowly enough that any difference in lighting or angle will be larger than any real difference in the skin, which is why this is worth the trouble with GHK-Cu and the RADIANCE blend.

For sleep: seven nights of lights-out time, when you think you fell asleep, how many times you woke, and a rating of how you felt on waking. The Insomnia Severity Index is a short, free alternative to an ad-hoc score, and a 2011 paper in Sleep established the change in score corresponding to a meaningful treatment response. A wearable supplements that and does not replace it — a 2021 study in Sleep tested seven consumer sleep-tracking devices against laboratory polysomnography in 34 adults and found most about as good as research actigraphy at telling sleep from wake, but specificity for correctly identifying the minutes you were actually awake ranged from 0.18 to 0.54, and devices did worse on disrupted nights. With DSIP and the SERENITY blend, the written record is the primary evidence.

A week of daily one-to-ten ratings written at the same time each day. Ratings made at the time beat ratings made in retrospect: the 2008 Annual Review of Clinical Psychology overview of ecological momentary assessment sets out why recalled symptom reports are distorted by current state and by the most extreme moment recalled. One line a day is enough, and it is the only measure here that becomes impossible to recover later. This applies to NAD+, MOTS-c, Semax and Selank, where no lab value corresponds to the thing you are trying to change.

If your prescription touches the growth-hormone axis or metabolic health, draw labs before the first dose rather than after. At-home blood testing covers hormone, metabolic and thyroid markers without a lab visit, which removes the usual reason people skip the baseline draw.

IGF-1 is the closest thing this category has to a hard number. It is produced largely in the liver in response to growth hormone and is stable enough across the day that a single draw is interpretable, which is why it is the marker used in practice rather than a growth hormone level.

Its use as a monitoring marker is not a wellness convention — it appears in approved product labeling. The approved tesamorelin label states that the drug stimulates growth hormone production and increases serum IGF-1, directs prescribers to monitor IGF-1 during therapy, and describes considering discontinuation in patients with persistent elevations above roughly three standard deviation scores, particularly where the response is not robust. The approved somatropin label directs adult dose titration using clinical response, adverse reactions and age- and sex-adjusted serum IGF-1, decreasing the dose where IGF-1 rises above the age- and sex-specific normal range. The Endocrine Society's 2011 clinical practice guideline on adult growth hormone deficiency, in the Journal of Clinical Endocrinology and Metabolism, uses the same marker for titration.

Note what those labels say. IGF-1 confirms that a growth-hormone-axis medication is producing a physiological effect and keeps that effect inside a safe range. It is not a measure of how good you feel, and no label claims it is. Pepti's compounded sermorelin, CJC-1295 / Ipamorelin and tesamorelin are not FDA-approved finished products and none of that labeling applies to them as products; what carries across is the physiology and the monitoring logic.

Two caveats. Assay standardization is imperfect — a 2011 consensus statement in Clinical Chemistry on growth hormone and IGF assays exists precisely because results differ between methods, so use the same laboratory for the follow-up draw. And the older published work on GHRH(1-29), including the 1992 and 1997 studies in the Journal of Clinical Endocrinology and Metabolism in older adults, reports IGF-1 rising in treated groups: the effect the marker is designed to detect, not a promise about outcome. IGF-1 LR3 is a separate case, an IGF-1 analog rather than something acting through your own growth hormone, so a serum result does not mean on IGF-1 LR3 what it means on sermorelin.

The trend is the measurement; any single weight is a sample of it. Four weeks is the shortest interval over which it is readable, which lines up with how semaglutide and tirzepatide are supplied and refilled.

Waist circumference belongs next to it, measured at the navel, at the same point in the day, tape snug and level at the end of a normal exhale. A 2020 consensus statement in Nature Reviews Endocrinology from the International Atherosclerosis Society and the International Chair on Cardiometabolic Risk argued for recording it alongside weight as a routine clinical vital sign, because it tracks abdominal fat in a way body weight alone does not. On semaglutide or tirzepatide, the waist frequently moves in weeks where the scale sits still.

Appetite is what these medications act on most directly, so it is a reasonable early read. Rate it daily and add one behavioral line: whether you finished the portion, went back for a second serving, thought about food between meals. Appetite entirely unchanged several weeks into an established maintenance dose is a concrete, reportable observation. HbA1c, fasting glucose, a lipid panel and liver enzymes are the labs most often followed alongside; HbA1c reflects roughly the preceding three months, so it is not a four-week measure.

For repair, the same three numbers weekly from the same test movement: pain out of ten, degrees of motion, load managed. Load tolerated moves first and is the least corruptible by mood, and where a physiotherapist is involved their assessment beats yours. That matters more across BPC-157, TB-500, the BPC-157 + TB-500 preparation and KLOW than anywhere else, because there are no large randomized human trials of BPC-157 or TB-500 for tendon and soft-tissue injury to tell you what should have happened by a given week.

Where the goal is training rather than one injury — the ASCEND and TITAN blends — the equivalent measure is a training log, because what changes is the weekly volume you can sustain rather than any single session. Judge ASCEND or TITAN on a month of logged sessions, never on how one felt.

For skin, photographs monthly in fixed conditions, month one against month three rather than against last week — the whole protocol for GHK-Cu and the RADIANCE blend. For sleep, the nightly written line with DSIP or the SERENITY blend, leaning on total sleep time and sleep onset from any device and discounting its stage estimates.

For energy, mood and cognition there is no blood marker at all. The daily rating plus a note of what you actually did that day is the whole instrument, and the note is the part that makes the rating interpretable: a 7 on a day you trained and worked a full day is not the same 7 as a 7 on a day off. Rate a task you already do regularly rather than a new one, because novel tests improve with practice and that looks like a treatment effect. It is the honest limit of self-measurement with Semax, Selank, NAD+ and MOTS-c.