2.57pooled death risk ratio, three steroid cohorts, 2026 meta-analysis
Evidence · Eras
The steroid era has mortality data, and the peptide era mostly has a blank
Steroids have decades of cohort studies on who died and who got heart disease. Peptides mostly have case reports and a lot of silence, so we line the two records up without naming anyone and without claiming silence means safe.
As of 1 November 2026, published cohorts link non-medical steroid use with higher mortality (pooled risk ratio 2.57 across three cohorts) and heart disease. BPC-157 has 1 case report and 0 trial-tagged papers, so the peptide harm record is a blank, and a blank is not safety.
Steroid figures are from the abstracts of five papers read at PubMed on 1 November 2026. Peptide counts were read 30 October 2026. All steroid studies are observational.
The short version
2.57pooled death risk ratiosteroid exposure, three cohorts, 2026 meta-analysis
3.0heart attack hazard ratioDanish users against matched controls
1BPC-157 case reportson PubMed; trials: 0
11melanotan II case reportsthe loudest peptide harm record we found
Era claims against the recordwhat gets said, and what the record shows, read 30 October 2026
They sayThe steroid stuff is all exaggerated.
Not that simple
Three cohorts pooled to a risk ratio of 2.57 for death, though the authors say confounding and a small evidence base rule out proving cause. See the receipt
They sayPeptides have no side effects.
Not cleared
Peptides mostly have no cohort data at all, which is not the same as no side effects. Case reports exist for some. See the receipt
They sayPeptides are safer than steroids.
Not a decision
Nobody has measured it. Safer than what, by which outcome, over how many years? See the receipt
What do the steroid-era studies actually show?
Three steroid-era studies, three different measureshigher than comparison groups, read 1 November 2026
Each bar is a different measure from a different design: a death risk ratio, an adjusted hazard ratio for heart attack, and a pooled death risk ratio. They are not directly comparable, and all three are observational, so none proves cause.
A Finnish cohort of 62 powerlifters (2000) found 8 deaths over 12 years (12.9%) against 3.1% in population controls, a risk ratio of 4.6.[1] The authors say use was strongly suspected, not confirmed.
A Danish registry study in Circulation (2025) followed 1,189 sanctioned users and 59,450 controls for about 11 years, with an adjusted hazard ratio for heart attack of 3.00 (95% CI 1.67 to 5.39).[2]
A 2026 systematic review and meta-analysis pooled three cohorts to a death risk ratio of 2.57 (95% CI 1.98 to 3.33), with a prediction interval of 0.47 to 13.87.[3] The authors warn that residual confounding and a small evidence base preclude causal inference.
A Swedish study of former elite power-sport athletes found about 45% excess mortality between ages 20 and 50, but no increase over the whole follow-up.[4] Suicide was 2 to 4 times higher at ages 30 to 50, and cancer deaths were lower.
What does the peptide era have on record?
What the peptide era has on recordPubMed counts, read 30 October 2026
Counts of papers tagged as trials or case reports on PubMed. A case report is not a rate of harm, and a count of zero means no data, not no harm.
BPC-157 has 226 papers, 0 tagged as a trial and 1 as a case report. Our BPC-157 check has the sources.
Melanotan II has 11 case reports. A 2017 dermatology review counted four case reports of melanomas emerging from existing moles during or after use, and said conclusive evidence linking them is lacking.[5]
MK-677 has 16 trial-typed papers and is in FDA's Category 2 for a potential heart-failure risk, citing a trial stopped early.[6][7] Our MK-677 check covers it.
Why does the peptide record look so quiet?
How harm data usually builds upthe peptide era is near the start
Cohorts need years, registries and large groups. For steroids they exist; for most peptides they do not exist yet, which is why the record looks quiet.
Cohort studies of harm need registries, time and large groups of users. For peptides those do not exist yet, so the record is thin.
That makes the comparison lopsided. Steroids are measured, and peptides are mostly unmeasured, not shown safe.
How to read a harm headline
1. Ask which design it is: a case report, a cohort, or a trial. They answer different questions.
2. Ask what it compared against, and for how long.
3. Ask whether use was confirmed or only suspected.
4. A blank record means unknown, not safe.
5. Take your own health questions to a licensed clinician.
What should you do with this?
If you saw a "steroids vs peptides" takeAsk what was measuredOne side has cohorts and the other has case reports. A fair comparison needs the same kind of data on both.
If you want the sourcesRead the abstracts, not the headlinesThe five PubMed records linked above give design, size and the authors' own cautions.
If this touches your own healthTalk to a licensed clinicianPopulation figures say nothing about one person, and none of this is medical advice.
How these figures were read
Method
The steroid-era figures are copied from the abstracts of PMIDs 10834358, 24033718, 39945117 and 42930597 and the metadata of 28266027, read at PubMed on 1 November 2026. We did not read the full texts, so subgroup details are not covered.
The peptide-era counts come from the same PubMed queries as our compound checks, read 30 October 2026. This page names no individual. A named case gets its own receipt only when a medical examiner, court or peer-reviewed source states the facts, and it says only what that source says.
References
↑ Pärssinen M, et al. Increased premature mortality of competitive powerlifters suspected to have used anabolic agents. Int J Sports Med, 2000. PMID 10834358. PubMed, read 1 November 2026.
↑ Anabolic-Androgenic Steroid Use, All-Cause Mortality, and Major Cardiovascular Outcomes: A Systematic Review and Meta-Analysis. Med Sci (Basel), 2026. PMID 42930597. PubMed, read 1 November 2026.
↑ Lindqvist AS, et al. Increased mortality rate and suicide in Swedish former elite male athletes in power sports. Scand J Med Sci Sports, 2014. PMID 24033718. PubMed, read 1 November 2026.
↑ ClinicalTrials.gov record NCT00128115, terminated, hip fracture, phase 2. clinicaltrials.gov, read 1 November 2026.
↑ FDA. Certain bulk drug substances for use in compounding that may present significant safety risks (content current as of 22 April 2026). fda.gov, read 30 October 2026.
FAQ
Do steroids really raise the risk of death?
Published observational cohorts link non-medical steroid use with higher death rates and with heart disease, and a 2026 meta-analysis pooled three cohorts to a risk ratio of 2.57. The authors say confounding and a small evidence base mean this does not prove cause.
Are peptides safer than steroids?
Nobody has measured it in the same way. Steroids have registry cohorts, and most peptides have only case reports, so there is no fair head-to-head.
Does no harm data mean peptides are safe?
No. It means no one has gathered the data. A blank record is unknown, and FDA still lists some peptide-style compounds in Category 2 for safety risks.
Was steroid use confirmed in these studies?
It varies. The Finnish powerlifter study says use was strongly suspected, not confirmed, while the Danish study used men sanctioned in an antidoping programme.
Why does this page not name anyone?
Because a study cohort is not a person. When we cover a named case, it gets its own receipt, cites the medical examiner, court or peer-reviewed source, and says only what that source says.
Bottom line
Steroid harms are measured in cohorts, with caveats the authors state themselves. Peptide harms are mostly unmeasured, and a quiet record is a gap in the data, not a safety result. Ask what kind of data any claim rests on, and take your own questions to a licensed clinician.