REF-1590 · Van Wagoner et al., 2017 · JAMA · Paper · PubMed · Free full text

Of 44 products sold online as SARMs and bought in 2016, only 23 (52%) contained a SARM, 17 (39%) contained another unapproved drug such as the growth hormone secretagogue ibutamoren, 4 (9%) held no active compound, 11 (25%) held substances not on the label and only 18 (41%) matched the labelled amount; from the abstract

Main caveat: 44 products, one country, one window in 2016, online sellers only; markets change.
limited evidenceMeasured on a mixed groupRead from the abstract only

Van Wagoner and colleagues bought products marketed online as selective androgen receptor modulators (SARMs), drugs not approved by the US Food and Drug Administration that are used to change appearance and performance, and measured what was in them.

Of 44 products, about half held a SARM; many held other unapproved drugs, some held nothing active, a quarter held substances missing from the label, and fewer than half held the labelled amount. The authors conclude that most products contained unapproved drugs and that many were inaccurately labelled.

Participants

No people: 44 products marketed and sold online as SARMs (dietary supplements and other products), United States, 2016.

Protocol

Sampling: Google searches in Chrome and Internet Explorer from 18 February to 25 March 2016 to find suppliers; products bought.
Analysis: April-August 2016, chain of custody, World Anti-Doping Agency-approved analytical procedures; identities and amounts compared with labels.

Key Results

FindingProducts (of 44)Share
Contained 1 or more SARMs (ostarine, LGD-4033 or andarine)2352%
Contained another unapproved drug (ibutamoren, GW501516, SR9009 among them)1739%
No active compound detected49%
Substances not listed on the label1125%
Amount matched the label1841%
Labelled amount differed substantially from what was found2659%

Practical Application

Can a "research chemical" bought online be trusted? Of 44 products sold online as SARMs in the US in 2016, only half contained a SARM, a quarter held substances not on the label and fewer than half matched the stated dose, so what is in the vial cannot be known from the label; this tested SARM products, not peptides, though an undeclared growth hormone secretagogue was among the findings.
Inference, not tested here: a product carrying an unlisted drug exposes a tested athlete to a positive test, and anyone to a dose they did not choose, without their knowing it.
SARMs are not approved by the US FDA and the other drugs found were also unapproved (the abstract's terms). Inference, not tested here: when the dose in the vial differs from the label in most products, the user cannot know what dose they took, let alone whether it is safe.
Practice, not from this paper: the base's supplement rule is to buy third-party-certified products (SYNTH-67); for products sold as SARMs or "research chemicals" the base's answer is not to use them.

Limitations

44 products, one country, one window in 2016, online sellers only; markets change.
Written from the abstract (the full text is in PMC, but Europe PMC does not list it as open access).
Chemistry only; no health outcomes.
How this entry was checked and changed
Why this level

a cross-sectional measurement study of fewer than 100 products (B+ needs 100 or more); laboratory chemistry against the label is a direct measure, not a marker