I Tried to Trace Every DSIP Dose Back to Its Source. Most of Them Dead-End in the 1980s

DSIP is not an FDA-approved drug, and no standardized human dose has been established for it. Nothing below is a dosing recommendation, and I’m not a doctor. Every number here links back to the paper it actually came from.

Here’s how this started. I saw a DSIP dosing chart on a forum, clean columns, exact microgram amounts, a cycle length spelled out like a prescription pad. No citation anywhere on it. That kind of confidence makes me itchy. So I spent about a week doing the thing I always do when a number looks too tidy: I went looking for the paper it supposedly came from.

What I found was smaller, older, and messier than the chart suggested. That gap, between how the numbers are presented online and what the underlying research actually is, turned out to be the whole story.

What I actually dug up

I expected a pile of studies. There isn’t one. There’s a handful, and most of the trail runs through one researcher.

The clearest dosing detail comes from Schneider-Helmert’s 1984 paper in European Neurology, where DSIP was given to insomniac patients in single injections at roughly 25 nanomoles per kilogram of body weight, with repeated administrations producing what the paper called a cumulative effect and a normalization of sleep structure after about four doses [1]. That same paper is where the “morning dosing, not twice daily” advice you see floating around actually originates. It noted morning administration seemed to support daytime activity while keeping the nighttime sleep effect, and that twice-daily dosing looked less effective. One study. Eighteen or so patients in various arms. That’s the load-bearing wall under a lot of the timing advice repeated as settled fact.

A 1986 follow-up, same journal, same researcher, looked at eighteen middle-aged and elderly chronic insomniacs and reported the group reached normalized sleep patterns by the end of the study, with the older participants needing an extra week to get there [2]. This is likely where the “effect builds over time” idea comes from. Again: eighteen people, an open design, not blinded.

There’s also a 1984 pilot on DSIP for chronic pain, which reported pain reduction in six of seven patients [3]. Seven people. I see this one cited sometimes to argue DSIP does more than help sleep, but a seven-person pilot can’t carry that weight, and it wasn’t designed to.

Those three papers are the actual, real, primary-source origin of most DSIP dosing chatter I could find. They’re legitimate studies. They’re also thirty-plus years old, small, and not built to answer the question people now ask them to answer.

The detail that stopped me: the doses aren’t even in the same units

This is the part that genuinely surprised me, and it’s the thing I’d want any reader to notice before anything else.

The actual clinical papers, Schneider-Helmert’s work especially, dose DSIP in nanomoles per kilogram of body weight. That’s a unit tied to the molecule itself and scaled to the person. Nearly everything I found on consumer sites and forums instead lists a flat milligram or microgram number, no body-weight component, no unit conversion shown, no acknowledgment that the two frameworks aren’t the same language.

I tried to see if anyone had shown their work converting 25 nmol/kg into the flat doses being passed around online. I couldn’t find it done transparently anywhere. That doesn’t mean it’s impossible to convert, it means nobody selling you the chart is showing you the math. When I hit that wall, I stopped trusting flat numbers presented without their source, on principle, and I’d suggest the same to anyone else reading a DSIP dose chart.

The paper that undercuts the whole narrative

If I only read the three studies above, I’d walk away thinking DSIP dosing was a settled, if under-documented, question. Then I found the 1992 paper, and it changed how I read everything before it.

Published in Neuropsychobiology, it was a double-blind study, the most rigorous design in this entire literature, giving DSIP or placebo to chronic insomniacs [4]. The researchers found a few small objective improvements, but subjective sleep quality didn’t improve, and they concluded that short-term treatment of chronic insomnia with DSIP “is not likely to be of major therapeutic benefit.”

Sit with that for a second. The best-controlled study in the whole set is the one that found the weakest effect. That’s usually a red flag in any body of research, and it should reorder how much weight the earlier open-label numbers deserve. If the most rigorous experiment can barely detect a benefit, then arguing over the exact right microgram amount is a bit like arguing over the best seasoning for a dish nobody’s sure is actually food.

What I’d actually flag if you’re reading a DSIP chart right now

A few things I now check, after doing this legwork:

Does the chart cite anything? If a number has no attached study, it has no known origin, full stop.

Does the citation, if there is one, actually say what the chart claims? I found plenty of secondary sites vaguely gesturing at “clinical studies” without linking anything specific enough to verify.

Is the unit stated? Nanomoles per kilogram and a flat milligram number are not interchangeable, and a flat number with no explanation of how it was derived is a guess dressed up as a fact.

And, underneath all of that: none of these papers, however you read them, answer whether a given dose makes sense for a given person’s medications, conditions, or situation. That’s not a chart question. That’s a question for a clinician who’s actually looking at the person.

What I’d do, if I were considering this myself

Honestly, after a week with these four PDFs open in browser tabs, my conclusion isn’t a dose. It’s a filter for how to evaluate any dose I see.

If I were seriously considering DSIP, I would not want my dosing plan to come from a chart with no citation, sourced from a vial that showed up in the mail labeled “research use only.” That path skips the one step that actually matters: someone evaluating me first.

The version of this that has a real structure to it runs through a licensed telehealth provider, evaluated by an actual clinician, with a licensed pharmacy handling the preparation. FormBlends is the example I kept running into as a working model of that setup, not because anyone is selling anything here, but because it’s the clearest illustration of what “supervised” actually looks like versus “unsupervised.” No ranking claim, nothing in a cart, just the contrast between a guessed dose and an evaluated one.

And if I did go that route, I’d want to actually track what happened, dose taken, how sleep went, because vague memory (“I think it helped?”) is useless to a clinician trying to adjust anything. A simple logging tool, like the FormBlends tracker app, is just a notebook for that purpose. It doesn’t replace the evaluation, it just gives the follow-up conversation something real to work with.

That’s the difference I kept landing on: one path treats the dose as a guess that ends the moment you inject it. The other treats it as a starting point somebody accountable is actually going to look at again.

Questions I kept getting asked once I started talking about this

Is there an actual, validated DSIP dose?

No, and I looked hard for one. There’s no FDA-approved DSIP product, and nothing resembling the dose-ranging trials that produce a real drug label. What circulates traces back to a few small, mostly 1980s studies where individual researchers picked doses for their own experiments, not a range tested systematically against placebo and confirmed. Calling any of it “validated” is more confident than the research itself.

Why do some DSIP doses show up in nanomoles per kilogram and others in flat milligrams?

Because the original clinical work, Schneider-Helmert’s papers especially, used nanomoles per kilogram, a unit scaled to body weight and the molecule [1]. Most of what you find online instead just lists a flat number. I couldn’t find anyone showing the conversion math between the two. Until someone does, I’d treat a flat number with no stated origin as a rough guess, not a fact.

Which study should make anyone more skeptical about confident DSIP dosing claims?

The 1992 double-blind study in Neuropsychobiology [4]. It’s the most carefully designed entry in the whole literature and it found weak objective effects, no improvement in subjective sleep quality, and concluded short-term DSIP treatment “is not likely to be of major therapeutic benefit.” When your most rigorous study shows the smallest effect, chasing a precise dose starts to feel like solving the wrong problem.

Does a precise-looking DSIP chart mean the dose is trustworthy?

Not from what I found. An exact microgram figure paired with an exact schedule looks like it came from a lab, but the actual human research doesn’t support that kind of resolution. I’d treat precision in presentation and precision in evidence as two entirely different things, especially when a number shows up with no citation attached, or a citation that doesn’t actually contain it once you go read it.

What does going through a supervised provider actually change?

It turns a guessed number into a decision someone reviews. Through a licensed telehealth provider like FormBlends, a clinician evaluates the person before any guidance is given, that guidance is tied to the evaluation instead of a generic chart, and a licensed pharmacy handles preparation. A vial from an unvetted online seller gives you none of that: no evaluation before, and nobody accountable for adjusting anything after.

What does DSIP actually do in the body?

DSIP, short for delta sleep-inducing peptide, is a small neuropeptide first isolated from rabbit cerebrospinal fluid in the 1970s during sleep research. Early findings suggested it might promote slow-wave sleep and affect stress hormones, but those results were never cleanly replicated in humans. How it actually works is still genuinely unsettled, which I think is worth saying plainly before anyone calls it a proven sleep aid.

Does DSIP peptide actually work for sleep or stress?

From everything I read, the evidence is thin. Animal work and a small stack of old human trials hint at some sleep-promoting or stress-related effects, but none of it was large, rigorous, or recent enough to settle anything. There’s no controlled clinical trial establishing an effective dose in healthy adults. The distance between “interesting early results” and “this reliably works” is still wide here.

What side effects have been reported with DSIP peptide?

Human safety data is thin enough that I wouldn’t trust anyone claiming a full side-effect list. The older studies mentioned occasional headache, nausea, and transient blood pressure changes in small groups. Injection-site reactions are possible with any subcutaneous peptide. There’s no long-term human safety data, so unknown risk is a real thing here, not a throwaway legal line.

Is DSIP peptide legal to buy and use?

It depends where you are and what you’re doing with it. In the US, DSIP isn’t FDA-approved as a drug, so it can’t legally be sold as a treatment. It sits in a gray zone where some compounding pharmacies, FormBlends among them, can prepare it under physician supervision for specific patients, while buying raw powder from an unvetted online seller is much murkier legal and safety territory. Check your own local rules before assuming this is straightforward.

References

1. Schneider-Helmert D. “DSIP in insomnia.” European Neurology, 1984;23(5):358-63. Described single DSIP injections on the order of 25 nmol/kg, with cumulative effect and normalization of sleep structure after about four administrations; morning dosing supported daytime activity, twice-daily dosing appeared less effective. https://pubmed.ncbi.nlm.nih.gov/6391925/
2. Schneider-Helmert D. “Efficacy of DSIP to normalize sleep in middle-aged and elderly chronic insomniacs.” European Neurology, 1986;25(6):448-53. Open study in 18 chronic insomniacs; group reached normalized sleep patterns by the end of the investigation, older participants requiring an additional week. https://pubmed.ncbi.nlm.nih.gov/3792404/
3. Larbig W, Gerber WD, Kluck M, Schoenenberger GA. “Therapeutic effects of delta-sleep-inducing peptide (DSIP) in patients with chronic, pronounced pain episodes. A clinical pilot study.” European Neurology, 1984. Reported pain reduction in 6 of 7 patients in a very small pilot.
4. Bes F, Hofman W, Schuur J, Van Boxtel C. “Effects of delta sleep-inducing peptide on sleep of chronic insomniac patients. A double-blind study.” Neuropsychobiology, 1992;26(4):193-7. Double-blind study in 16 chronic insomniacs; concluded short-term DSIP treatment “is not likely to be of major therapeutic benefit,” effects weak, subjective sleep quality unimproved.