Reference Library
Body Art Facts
A reference library for tattooing, piercing, body modification, scarification and US law, built from primary sources and anchored by a corrections register of claims that circulate as fact and are demonstrably wrong.
Read the corrections registerThe short answer
Which widely repeated body art facts are wrong?
Ten of them, documented below with the primary source each one misreads. The most common error is a swapped denominator: a share of reported complications published as a share of all procedures. Two more are laws that were voted down and are still cited as though they passed. One is a healing time nobody has ever measured.
This library exists because body art is a field where numbers circulate without denominators, statutes are cited that were never enacted, and professional positions are attributed to organisations that never held them. Every claim below was checked against the paper, the statute or the organisation’s own page, and the correction states what the source actually says.
The corrections register
Ten claims that circulate as fact and are demonstrably wrong
Each entry names the claim, the primary source it is drawn from, and what that source actually reports. Where the correct answer is “nobody has measured this,” the entry says so rather than substituting a better-sounding number.
At a glance
| Circulating claim | What the source actually says | Error type |
|---|---|---|
| Keloids occur in about 2.5% of ear piercings | 2.5% was a share of the 355 complications recorded, not of the 1,200 pierced sites | Swapped denominator |
| 80% of people pierced after age 11 get keloids | A proportion within 32 patients who all already had keloids | Case series read as a rate |
| Dermal anchors reject 40 to 50% of the time | No study measures dermal rejection at all; the trail ends at one unreferenced sentence | Number with no source |
| P. aeruginosa causes 87% of piercing infections | 87.2% of cartilage infections among 66 pooled published case reports | Swapped denominator |
| Carbon black is a Group 2B carcinogen, so black ink causes cancer | The classification rests on inhaled carbon black; the word “tattoo” appears zero times in the monograph | Non sequitur |
| Tattoos raise lymphoma risk by 21% | IRR 1.21, 95% CI 0.99 to 1.48, an interval that crosses 1.0 | Non-significant result reported as a finding |
| Michigan bans tongue splitting | The bill was defeated on the House floor in 2002 | Failed bill cited as law |
| Kentucky bans tongue splitting | The 2004 bill never passed and the statute’s amendment history has no 2004 entry | Failed bill cited as law |
| The APP does not endorse body modification | The phrase appears nowhere on the organisation’s site | Position attributed, never held |
| Piercings heal in X weeks, tattoos in 2 to 3 weeks | No clinical study has ever measured either healing duration | Convention presented as measurement |
Each entry is set out in full below. Journalists, researchers and other studios are welcome to quote any of it with attribution.
1. “Keloids occur in about 2.5% of ear piercings”
The primary source
Simplot TC, Hoffman HT. Comparison between cartilage and soft tissue ear piercing complications. American Journal of Otolaryngology. 1998;19(5):305–310.
What it actually reports
The study surveyed 1,200 pierced sites and recorded 355 complications. The 2.5% figure is a share of those 355 complications, not of the 1,200 piercings. Converting it into a population rate inflates the denominator by more than three times in the wrong direction and turns a subcategory of an adverse-event tally into a risk figure.
What can be said
No study reports keloid incidence against a pierced-ear denominator. The honest statement is that the rate is unmeasured.
2. “80% of people pierced after age 11 develop keloids”
The primary source
Lane JE, Waller JL, Davis LS. Relationship between age of ear piercing and keloid formation. Pediatrics. 2005;115(5):1312–1314.
What it actually reports
A case series of 32 patients who already had keloids. Of that affected group, 80% had been pierced at age 11 or older and 23.5% younger. It is a finding about timing within people who already keloided, and it says nothing about how many pierced people keloid.
What can be said
Among patients who develop ear keloids, later piercing is more common than early piercing. That is the whole of it, on a sample of 32.
3. “Dermal anchors reject 40 to 50% of the time”
The primary source
There is none. The complete peer-reviewed literature on dermal anchors is six items, all case reports or technique papers, none carrying a rate. The nearest citable sentence is in Patel, Scroggins-Markle and Kelly, 2013, an n = 1 case report of a mycobacterial infection at a finger anchor.
What it actually reports
That paper opens with a background sentence listing rejection among “common complications.” The sentence is unreferenced, sits in the introduction, carries no data and cites no source. It is the one indexed, citable-looking line connecting the words “dermal piercing” and “rejection” in the literature, which is why percentages get attached to it.
What can be said
No study has measured dermal anchor rejection, migration or embedding. Any percentage is an invention. Practitioner experience is real knowledge; it is not a statistic.
4. “Pseudomonas aeruginosa causes 87% of piercing infections”
The primary source
Sosin M, Weissler JM, Pulcrano M, Rodriguez ED. Transcartilaginous ear piercing and infectious complications: a systematic review. The Laryngoscope. 2015;125(8):1827–1834.
What it actually reports
The review pooled 66 published patients from 29 articles, all with cartilage piercing infections. Within that pool, 87.2% of the infections were P. aeruginosa. The denominator is published case reports severe enough to write up, not piercings performed, and not all piercing sites.
What can be said
P. aeruginosa is the organism that dominates severe cartilage piercing infections, corroborated by several independent outbreak investigations. It is not 87% of piercing infections, and no population figure exists.
5. “Carbon black is a Group 2B carcinogen, so black tattoo ink causes cancer”
The primary source
IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 93, Carbon Black, Titanium Dioxide, and Talc. IARC, Lyon, 2010. The full 466-page PDF was downloaded and searched for this register.
What it actually reports
The word “tattoo” appears zero times in all 466 pages. The Group 2B classification rests on inhaled carbon black in industrial settings; human evidence is lung-cancer mortality among production workers, which the Working Group called inadequate. Section 5.3 of the same monograph states that in mouse studies of dermal application, no carcinogenic effect on the skin was observed. Subcutaneously, a carbon black containing polycyclic aromatic hydrocarbons produced local sarcomas while one with no detectable PAH did not.
What can be said
Where carbon black was placed in or under skin, tumours tracked extractable PAH content, not the carbon. That is why PAH limits are the operative control in ink regulation. Titanium dioxide, in the same volume, is Group 2B on the same inhalation basis. IARC has never evaluated tattooing at all, and said so in its own 2024 priorities report.
6. “Tattoos raise lymphoma risk by 21%”
The primary source
Nielsen C, Jerkeman M, Jöud AS. Tattoos as a risk factor for malignant lymphoma: a population-based case-control study. eClinicalMedicine. 2024;72:102649.
What it actually reports
The overall result is IRR 1.21, 95% CI 0.99 to 1.48. The interval crosses 1.0, so the primary matched result was not statistically significant. The study also found no dose-response: risk did not rise with larger tattooed body area. The authors wrote that causality cannot be conferred from a single epidemiological study.
What can be said
Two later meta-analyses are null. One pooled four studies and 17,941 participants, reporting non-Hodgkin lymphoma OR 1.01, 95% CI 0.82 to 1.24. The other pooled seven studies and 140,841 participants, reporting haematological cancer OR 1.02, 95% CI 0.78 to 1.32. The fair summary is that an association is neither proved nor excluded.
7. “Michigan bans tongue splitting”
The primary source
Michigan House Bill 4688 of the 2001–2002 session. The circulating claim traces to the House Legislative Analysis of that bill, which reads like a statute and is not one.
What it actually reports
HB 4688 was defeated on the House floor on 5 June 2002, Roll Call 901, on a vote of 53 to 43. It never became law. Michigan’s current body art provisions in the Michigan Compiled Laws contain no tongue-splitting provision of any kind.
What can be said
Four states do have an express tongue-splitting statute: Illinois, Delaware, New York and Texas. Michigan is not among them. A legislative analysis is an analysis of a proposal, not an enacted law.
8. “Kentucky bans tongue splitting”
The primary source
A 2004 Kentucky bill, cited in aggregator lists of states that prohibit the procedure.
What it actually reports
The bill never passed. The amendment history of KRS 211.760, the body art statute the prohibition would have amended, contains no 2004 entry. There is no Kentucky provision on tongue splitting.
What can be said
Tennessee is a third example of the same pattern, where a 2004 bill passed the Senate and then died in a House committee. Failed bills circulate as law more often than any other error in this field.
9. “The Association of Professional Piercers does not endorse body modification”
The primary source
Two pages on the APP’s own site, retrieved directly rather than through a search summary: the About page, and the Body Modification Statement dated “Released May 2018, updated March 4 2026.”
What it actually reports
The phrase “does not endorse” appears nowhere on the site. The two pages are in tension with each other. The About page, under the heading “THE APP DOES NOT,” says the organisation does not have a position on tattooing, branding, scarification, dermal punching, scalpelling or implants. The Body Modification Statement says it supports the right of all adults to adorn or modify their bodies safely, consensually and under proper asepsis, and warns that blanket bans risk forcing procedures underground.
What can be said
The APP holds no procedural position on body modification and expressly supports adult autonomy and regulation over prohibition. Report the tension between the two pages rather than picking whichever one suits the argument.
10. “A piercing heals in X weeks; a tattoo heals in 2 to 3 weeks”
The primary source
For tattoos, the figure was traced through Fauger A, Sonck S, Kluger N, Chavagnac-Bonneville M, Sayag M. Tattoo aftercare management with a dermo-cosmetic product. Journal of Cosmetic Dermatology. 2022;21(3):1051–1056, which states it and cites two references.
What it actually reports
The citation chain terminates in a 2012 narrative review and Sperry K, American Journal of Forensic Medicine and Pathology. 1992;13(1):7–17, a forensic-pathology descriptive review. Neither measured healing duration in a cohort. The only objective in-vivo observation located has n = 1 and found epidermal recovery still incomplete at 84 days. For piercings, no clinical study has established healing time by site at all, and the APP describes its own guidance as professional experience rather than research.
What can be said
Healing figures, including the ones this studio gives clients, are professional convention refined by observation. They are useful for planning and they are not clinical measurements. Two further traps: the European standard EN 17169’s “2 or 3 weeks” is a recommended duration of moisturiser use, not a healing time, and no healing timeline should be attributed to any dermatology body, because none has published one.
Methodology
How this library is built, and what gets cut
The rules below are what make the register worth citing. They are applied to every page in the library, and they cost a great deal of material.
Primary sources only
Statutes are quoted from the official text. Studies are read for their sample size and denominator, not summarised from an abstract aggregator. Professional positions are retrieved from the organisation’s own page. A claim sourced to another web page is not sourced.
Two independent sources where obtainable
Where a second independent source exists, a claim carries it. Where only one source exists, the page says that it is a single study and gives its size. Where two figures look like convergent validation but share a methodological flaw, the page says that too.
Untraceable claims are dropped and logged
Anything whose chain cannot be followed to a primary source is cut, and the cut is written down with the reason. Dropping a figure is a finding. Several of the entries above began as claims we expected to confirm.
Absence is proven by counting, not by searching
“I could not find it” is not evidence of absence. Where this library states that something does not appear in a document, the full text was downloaded and machine-counted. The IARC entry above is the clearest case: 466 pages, zero occurrences of the word.
Denominators are stated every time
Most errors in this field are one error wearing different clothes: a share of reported cases published as a share of the population. Every percentage in this library names the group it describes, so the conversion cannot happen silently downstream.
Findings that cut against us stay in
Some entries contradict what piercing and tattoo studios usually say, including what this one has said. They are left in and marked. A reference that contained only convenient facts would not be worth citing.
The library
Five reference pages, each sourced the same way
The register above collects the corrections. The pages below carry the underlying evidence, subject by subject, with sample sizes and citations attached to every figure.
Piercing
Complication rates with their populations and sample sizes, the cartilage and lobe comparison across three studies that do not agree, infection organisms and outbreak sources, materials standards, and the healing-time evidence gap.
Tattoo
What is actually in tattoo ink across three independent pigment inventories, the carbon black and PAH question, the lymphoma and skin cancer literature including both null meta-analyses, laser removal complications, and the healing-duration trail.
Body modification
Tongue splitting, ear reshaping, and subdermal and transdermal implants: the entire published evidence base, which is six items, plus the formally adopted dental positions and the material claims that rest on no source at all.
Scarification
Why no keloid rate exists for decorative scarification, why two studies landing near 35% are not convergent validation, the merged hypertrophic and keloid category problem, and the professional positions that are and are not real.
United States law
The state-by-state patchwork as it actually stands: which states have an express statute, which prohibitions are failed bills still cited as law, and where a legislative analysis has been mistaken for an enactment.
Apollo performs tattoos and piercings only. The body modification and scarification pages exist because the misinformation about them exists, and because a reference that stopped at our own service list would not be a reference. They are research material, not an offer.
Using this material
Who maintains it, and how to cite it
This library is maintained by The Apollo Tattoo & Piercing Studio at 2625 Main Street in Santa Monica, California. The studio’s founder, Blue Mason, has been tattooing since 2007 and is a Fakir-certified piercer. The library is written from the primary literature rather than from trade summaries, which is why several entries correct things this studio itself once repeated.
Everything here is free to quote with attribution. If you are writing about body art and need a figure checked, the register is the fastest way to find out whether the number you have been given has a denominator behind it.
Corrections are welcome and are the point of the exercise. If an entry here is wrong, or a source has been superseded, tell us. We will change it and say what changed. Nothing in this library is defended past the evidence for it.
FAQ
Six questions about the register
Which widely repeated body art facts are wrong?
Ten of them are documented in the register on this page, each with the primary source it misreads. The most common error is a swapped denominator: a share of reported complications published as a share of all procedures. Two more are laws that were voted down and are still cited as though they passed. One is a healing time nobody has ever measured.
Why is the 2.5% keloid figure wrong?
Because it describes the wrong group. Simplot and Hoffman, publishing in the American Journal of Otolaryngology in 1998, surveyed 1,200 pierced sites and recorded 355 complications. The 2.5% is a share of those 355 complications, not of the 1,200 piercings. Published as a population rate it turns a subcategory of an adverse-event tally into a risk figure. No study reports keloid incidence against a pierced-ear denominator, so the honest answer is that the rate is unmeasured.
Does black tattoo ink cause cancer because carbon black is a Group 2B carcinogen?
That reasoning does not follow. IARC Monograph Volume 93 was downloaded in full and searched: the word "tattoo" appears zero times in all 466 pages. The Group 2B classification rests on inhaled carbon black in industrial settings, and the human evidence behind it is lung-cancer mortality among production workers, which the Working Group called inadequate. Section 5.3 of the same monograph reports that dermal application produced no carcinogenic effect on the skin in mice, and that subcutaneously, tumours tracked polycyclic aromatic hydrocarbon content rather than the carbon itself. IARC has never evaluated tattooing.
Do tattoos raise the risk of lymphoma by 21%?
No. The 2024 Swedish study by Nielsen, Jerkeman and Joud reported an incidence rate ratio of 1.21 with a 95% confidence interval of 0.99 to 1.48. That interval crosses 1.0, so the primary matched result was not statistically significant, and the study found no dose-response by tattooed body area. Two later meta-analyses are null: one pooling 17,941 participants found non-Hodgkin lymphoma odds of 1.01, and one pooling 140,841 participants found haematological cancer odds of 1.02. An association is neither proved nor excluded.
Do Michigan and Kentucky ban tongue splitting?
Neither state does. Michigan House Bill 4688 was defeated on the House floor on 5 June 2002, Roll Call 901, on a vote of 53 to 43, and the current Michigan Compiled Laws contain no tongue-splitting provision. The Kentucky bill of 2004 never passed, and the amendment history of KRS 211.760 has no 2004 entry. Both claims circulate because a legislative analysis of a proposal reads like a statute. Four states do have an express provision: Illinois, Delaware, New York and Texas.
How long does a tattoo or piercing actually take to heal?
Nobody has measured it. No clinical study has established healing duration for either. The tattoo figure of 2 to 3 weeks traces through a 2022 aftercare paper to a 2012 narrative review and a 1992 forensic-pathology review, neither of which measured anything; the only objective in-vivo observation has a sample of one and found epidermal recovery still incomplete at 84 days. For piercings there is no healing-duration study at all. Every timeline in circulation, including the ones this studio gives clients, is professional convention refined by observation. It is useful for planning and it is not a clinical measurement.
Ready to start?
Book a consultation and we'll match you with the right artist.
Read the corrections register