Tattoos and body piercings in the United States: a national data set.

Abstract

Little is known about the prevalence and consequences of body art application.
Our aim was to provide US tattooing and body piercing prevalence, societal distribution, and medical and social consequence data.
Random digit dialing technology was used to obtain a national probability sample of 253 women and 247 men who were 18 to 50 years of age.
Of our respondents, 24% had tattoos and 14% had body piercings. Tattooing was equally common in both sexes, but body piercing was more common among women. Other associations were a lack of religious affiliation, extended jail time, previous drinking, and recreational drug use. Local medical complications, including broken teeth, were present in one third of those with body piercings. The prevalence of jewelry allergy increased with the number of piercings. Of those with tattoos, 17% were considering removal but none had had a tattoo removed.
This was a self-reported data set with a 33% response rate.
Tattooing and body piercing are associated with risk-taking activities. Body piercing has a high incidence of medical complications.

Full text

Page 1

REPORT
Tattoos and body piercings in the United States:
A national data set
Anne E. Laumann, MBChB, MRCP(UK),* and Amy J. Derick, MD
Chicago, Illinois
Background: Little is known about the prevalence and consequences of body art application.
Objective: Our aim was to provide US tattooing and body piercing prevalence, societal distribution, and
medical and social consequence data.
Methods: Random digit dialing technology was used to obtain a national probability sample of 253 women
and 247 men who were 18 to 50 years of age.
Results: Of our respondents, 24% had tattoos and 14% had body piercings. Tattooing was equally com-
mon in both sexes, but body piercing was more common among women. Other associations were a
lack of religious affiliation, extended jail time, previous drinking, and recreational drug use. Local medical
complications, including broken teeth, were present in one third of those with body piercings. The
prevalence of jewelry allergy increased with the number of piercings. Of those with tattoos, 17% were
considering removal but none had had a tattoo removed.
Limitations: This was a self-reported data set with a 33% response rate.
Conclusion: Tattooing and body piercing are associated with risk-taking activities. Body piercing has
a high incidence of medical complications. ( J Am Acad Dermatol 10.1016/j.jaad.2006.03.026.)
T
women sported tattoos,1and body piercings were
mainly in the soft part of the ear lobes of women.
Duringthe1940smilitarypersonneloftenhadtattoos
with patriotic designs, together with hearts and the
names of loved ones. With the advent of peace,
the popularity of tattoos decreased, although they
were still seen in close-knit group situations. They
becameassociatedwith
here has been an increase in the placement
of body art involving puncturing of the skin
in recent years. In the past more men than
marginalized groups,
signaling time spent in jail, ‘‘punk’’ status, member-
ship in a motorcycle gang, or a traveling circus. In
more recent times, images have become increasingly
eclectic and the practice has become mainstream.
Body piercing has been practiced in almost every
society and has usually been confined to the ears,
mouth, and nose. It is often related to ritual ceremo-
nies and religious rites.2In North America soft ear
lobe piercing for the secure insertion of earrings was
already commonplace in the early 1970s.3Recently,
piercing of other parts of the body has become
popular. Little is known about the current preva-
lence, societal distribution, and the medical and
social consequences of these activities.4This study
is an attempt to quantify some of these issues.
METHODS
Afteraseries of pretests, trained interviewers from
the Public Opinion Laboratory at Northern Illinois
University administered a telephone survey to 500
respondents aged 18 to 50 years between February
11 and April 18, 2004.
The respondents were selected using random
digit dialing (RDD) to obtain a national probability
sample within the 48 contiguous United States with
From the Section of Dermatology, University of Chicago.
*Current affiliation: Department of Dermatology, Northwestern
University; data collected while in the Section of Dermatology,
University of Chicago; analysis started while at the University of
Chicago and continued while at Northwestern University.
Supported by the Portes Center administered through the Insti-
tute of Medicine of Chicago and the Sage Foundation.
Conflicts of interest: None identified.
Accepted for publication March 15, 2006.
Reprint requests: Dr Anne Laumann, 676 N St Clair Street, Suite
1600, Chicago, IL 60611. E-mail: a-laumann@northwestern.edu.
Published online June 14, 2006.
0190-9622/$32.00
ª 2006 by the American Academy of Dermatology, Inc.
doi:10.1016/j.jaad.2006.03.026
1
ARTICLE IN PRESS

Page 2

Table I. Distribution and associations of tattoos and body piercings
Total (No.)*
100% (500)
Has a tattoo
24% (120)
Has ever had a
body piercing
14% (72)
Has had any
body art
30% (151)
Has had both
tattoo and body
piercing 8% (41)
Gender
Women
Men
51% (253)
49% (247)
22% (56)
26% (64)
21% (52)y
8% (20)y
33% (83)
28% (68)
10% (25)
7% (16)
Ethnicity
Black/African American
White/Cauacasian
Other
12% (57)
79% (392)
10% (47)
28% (16)
22% (86)
36% (17)
7% (4)
14% (54)
30% (14)z
33% (19)
28% (109)
47% (22)z
2% (1)
8% (31)
19% (9)
Hispanic ancestry
Hispanic
Non-Hispanic
9% (45)
91% (445)
38% (17)y
23% (102)y
27% (12)y
13% (59)y
47% (21)y
29% (128)y
18% (8)
7% (33)
Marital status
Married/engaged
Living with a partner
Divorced/separated
Widowed
Never married
51% (258)
6% (29)
12% (60)
1% (6)
27% (140)
20% (52)
41% (12)
27% (16)
50% (3)
25% (35)
11% (29)
17% (5)
12% (7)
0
22% (31)
25% (64)
48% (14)
30% (18)
50% (3)
36% (50)
7% (17)
10% (3)
8% (5)
0
11% (16)
Income
\$40K
$40K-$75K
[$75K
32% (161)
25% (124)
23% (116)
32%(52)y
24% (30)y
19% (22)y
21% (33)
14% (17)
14% (16)
39% (63)
30% (37)
28% (32)
14% (22)
8% (10)
5% (6)
Religion§
Religious affiliation:
No religious affiliation
64% (320)
34% (169)
19% (61)y
33% (55)y
11% (35)y
20% (34)y
25% (79)y
40% (67)y
5% (17)y
13% (22)y
Year of birth
1953-1963
1964-1974
1975-1986
36% (180)
34% (170)
28% (140)
15% (27)y
24% (41)y
36% (50)y
2% (4)y
13% (22)y
32% (45)y
15% (27)y
32% (54)y
48% (67)y
2% (4)
5% (9)
20% (28)z
Politics
Republican
Democrat
Independent
Other
31% (153)
28% (141)
20% (102)
14% (63)
22% (33)
24% (34)
27% (27)
29% (18)
8% (12)
16% (22)
19% (19)
18% (11)
25% (39)
31% (44)
35%(36)
33% (21)
4% (6)
9% (12)
10% (10)
13% (8)
Militaryk
Military experience
No military experience
10% (49)
89% (447)
37%( 18)y
23% (103)y
12% (6)
15% (66)
37% (18)
30% (133)
12% (6)
8% (35)
Jail time
$3 days or more
\3 days
10% (50)
90% (442)
58% (29)y
20% (90)y
26% (13)y
13% (58)y
62% (31)y
27% (118)y
22% (11)y
7% (30)y
Alcohol consumption
Never drank
Currently drink
Used to drink{
27% (133)
60% (295)
13% (63)
14% (19)y
26% (77)y
37% (23)y
6% (8)y
17% (51)y
19% (12)y
17% (23)y
34% (99)y
43% (27)y
3% (4)y
10% (29)y
13% (8)y
Ever used recreational drugs
Yes
No
31% (148)
69% (330)
38% (56)y
18% (59)y
22% (32)y
12% (39)y
45% (67)y
24% (78)y
14% (21)y
6% (20)y
J AM ACAD DERMATOL
2 Laumann and Derick
ARTICLE IN PRESS

Page 3

approximately equal numbers of men and women.
RDD allows for very wide coverage of an area, as it
includes numbers for households that have recently
moved, and for residents with unpublished num-
bers. Out of the initial bank of telephone numbers,
6416 were available for random drawing. Calling
attempts were made on different days and at differ-
ent times to try to reach respondents who had as
many different attitudes, habits, and characteristics
aspossible.Callsonweekdayswereoftenmadeafter
4:30 PM and late into the evening in order to reach
respondents in the western time zones.
Nineteen
(19,313) calls were made to complete 500 (Ncomplete)
interviews. Eighteen hundred sixty-one numbers did
not turn into completed interviews for the following
reasons: business numbers (217), nonworking num-
bers (663); fax/modem pager (298), group home (3),
no respondent meeting eligibility (554), language
barrier (99), physical/mental impairment (11), un-
available during period of interviewing (16). Sixteen
hundred thirty-two (1632) (Nrefusals) numbers were
labeled as refusals as follows: absolute refusal (761),
multiple attempts to reach respondent (52), multiple
thousandthreehundred thirteen
Table I. Cont’d
Total (No.)*
100% (500)
Has a tattoo
24% (120)
Has ever had a
body piercing
14% (72)
Has had any
body art
30% (151)
Has had both
tattoo and body
piercing 8% (41)
Education (respondents aged $24 y)
Total (No.)
Did not complete high school
Completed high school
Some college
Bachelors’ degree
Graduate school
100% (427)
10% (43)
20% (85)
30% (126)
26% (111)
15% (62)
23% (97)
40% (17)y
29% (25)y
25% (31)y
14% (16)y
14% (8)y
12% (52)
19% (8)
17% (14)
12% (15)
8% (9)
10% (6)
28% (118)
42% (18)y
33% (85)y
29% (36)y
23% (25)y
18% (11)y
7% (31)
16% (7)
13% (11)
8% (10)
0
5% (3)
Numbers in parentheses are numbers of respondents within each variable.
*Values in column 1 represent percentages of total for each group. Numbers may not add up to 100% in each group because of rounding
and occasional absent responses. Each percentage in columns 2, 3, 4, and 5 is the percentage for that variable of the total for the
characteristic in column 1.
yP \.05 between all adjacent cells within the same group.
zP\.05 related to the immediately superior cells in the same group, but the two cells above are not significantly different from each other.
§The question was ‘‘Regardless of whether you now attend any religious services do you think of yourself as part of a particular
denomination?’’
kThis is a combination of two questions: ‘‘Have you ever been or are you currently on active duty in the military, and are you a member of
the military reserve or National Guard?’’
{If the respondent denied drinking alcohol, the next question was: ‘‘Have you ever been a drinker?’’
Fig 1. Percentage of respondents tattooed or body pierced according to year of birth.
J AM ACAD DERMATOL
Laumann and Derick3
ARTICLE IN PRESS

Page 4

attempts at the household (578), multiple attemptse
busy (16), answering machine at known residence
(27), respondent claimed to have already completed
survey (14), pending appointment at end of survey
calling time (168), invalid/partial completes (16).
It was unknown whether 524 (Nunknown) numbers
belonged to a residence as follows: technical barrier
(15); multiple attempts made, number never an-
swered (340); answering machine (169). A number
was considered ‘‘dead’’ if after 8 attempts it did not
turn into a refusal, complete or unqualified number.
Respondents who initially refused to participate
were called back; 68 cases (14% of 500) became
completed interviews on callback.
The introduction of the survey follows: ‘‘Hello,
my name is ____________ and I’m calling from the
Public Opinion Laboratory on behalf of physicians at
the University of Chicago. We are doing a national
health study and I would like to ask you a few
questions.’’ So, the refusals were not related to the
body art content of the questionnaire.
Thecalculatedresponserate(RR3)was33%.5This
was derived as follows:
RR3 ¼
Ncomplete
Ncomplete1e?Nrefusals1Nunknown
where e (=48%) is the estimated proportion of
cases of unknown eligibility that are eligible. This
was estimated by using 2000 census data.
?
The average duration of the interviews was 11
minutes. If the respondent had no tattoos or body
piercings it was 8 minutes and if the respondent had
both it was 24 minutes.
The survey has a 4.4% margin of error. The
statistical analyses were performed using SPSS for
Windows, version 13.0 (SPSS, Chicago, Ill). Propor-
tions were compared using the two-tailed chi-square
test. Fisher’s exact test was used for rare responses.
The Public Opinion Laboratory is a full-service
survey research center in DeKalb, Illinois. The Insti-
tutional Review Boards of both the University of
Chicago and Northern Illinois University approved
the research instrument.
RESULTS
Respondents were equally distributed throughout
the age range (median and mean ages, 36 years
[standard deviation 9.44 years]). They were repre-
sentative of the nation according to race, Hispanic
ancestry, marital status, household size, median
household income, educational status, and religious
affiliation.(TableI,column1).Therewas anincrease
in the total number of people with or considering
getting body art (tattoos and body piercings) in each
Table IIA. Position and size of tattoos
Tattoo site and size
Male
(n = 64)
Female
(n = 56)
P
value
Face
Neck
Chest
Arms
Hands, fingers
Abdomen
Upper back, shoulder
Lower back
Buttocks
Hips
Legs
Ankle
Feet, toes
At least one exposedy
Only covered tattoos
Size is a palm or less
More than a palm
0 (0)*
5 (8)
11 (17)
48 (75)
9 (14)
3 (5)
22 (34)
3 (5)
2 (3)
1 (2)
13 (20)
1 (2)
1 (2)
57 (89)
7 (11)
35 (56)
28 (44)
2 (4)
2 (4)
11 (20)
4 (7)
1 (2)
7 (13)
14 (25)
10 (18)
3 (5)
5 (9)
8 (14)
15 (27)
2 (4)
27 (48)
29 (52)
45 (80)
11 (20)
NS
NS
NS
\.001
NS
NS
NS
.03
NS
NS
NS
\.001
NS
\.001
\.001
.003
.003
NS, Not statistically significant.
*Number (%) in category.
yExposed tattoos are on the face, neck, arms, hands, fingers, legs,
ankles, feet, and toes. These percentages did not vary by race.
Table IIB. Position of piercings: Total piercings by
site, respondents with soft ear lobe piercings, and
respondents with body piercings
MaleFemaleP value
Total piercings by site*
Part of jewelry in mouth
Ear (not soft ear lobe)
Nose
Belly button
Eyebrow
Nipple
Genitals
Soft ear lobe stretched
No. of respondents with
soft earlobe piercingsy
At least one soft
earlobe
Soft earlobes only
Respondents with BPsz
Had had [1 BP
At least one visible BP§
Covered BP only
21
28
5
3
9
6
5
3
19
66
13
29
5
1
2
0
47 (19)124 (49)
\.001
28 (11) 76 (30)
\.001
14 (70)
19 (95)
1 (5)
26 (50)
40 (77)
12 (23)
NS
NS
NS
BP, Body piercing; NS, not statistically significant.
*All piercings in a given location, including ones that have closed
and been repierced.
yMale respondents (n = 247); female respondents (n = 253);
number in parentheses is percentage.
zMale respondents (n = 20); female respondents (n = 52).
§Visible BPs include those in the hard cartilage of the ear, nose,
eyebrow, face, lip, or tongue.
J AM ACAD DERMATOL
4 Laumann and Derick
ARTICLE IN PRESS

Page 5

successive age group. (Table I and Fig 1). The start of
the modern fashion occurred in the early 1990s and
hascontinuedtoincrease.Tattooingwashappeningas
far back as 1969, when our first tattooed respondent
wasonly16,butbodypiercingwasnotseenuntil1982.
Tattooing
Three hundred forty-four tattoos were present
among 120 (24%) respondents, 60 of whom had
multiple tattoos with a range of 1 to 60. Tables I, IIA
and IIB, and IIIA and IIIB show additional data and
associations. An additional 104 (21%) nontattooed
respondents had considered tattoo placement.
(Fig 1). Younger respondents were more likely to
have multiple tattoos than the older cohorts. Thirty
percent of women compared with 11% of men had
obtained their first tattoo when they were older than
30 years of age. Overall, 65% of those with a tattoo
had obtained their first tattoo by 24 years of age.
Tattooedindividualsweremorelikelytohaveaclose
friend and/or family member with a tattoo than
nontattooed individuals (90% vs 63%; P\.001).
Tattooed respondents were less likely to claim a
religious affiliation than the nontattooed. Younger
people were less likely to claim a religious affiliation
than older people, but these variables behaved
independently with respect to their association
with the prevalence of tattooing.
To evaluate the educational status of those who
had body art, welooked only atrespondents aged 24
years and older so that they had had a chance to
complete college and to start graduate school. At all
levels of education, the prevalence of tattoos was
Table IIIA. Associations of tattooing related to the age of first application before or after age 18 years
Tattoo associations
First tattoo \18 y,
No. (%)
First tattoo $18 y,
No. (%)
Fisher exact
test (2-sided)
Total, No. (%)
Medical problems within the first 2 weeks*
Social/work problems due to tattoo
Did not finish high school
At least some graduate school
Jailed for $3 days
Drunk at time of first tattoo
Ever taken recreational drugs
High at time of first tattoo
First tattoo placed in a tattoo parlor
Professional tattoo gun for first tattoo
Professional artist made first tattoo
A friend/acquaintance made first tattoo
Had at least one tattoo done outside a studio
Ever considered having a tattoo removed
Ever considered having a tattoo covered
Ever had a tattoo covered
Has plans to get another tattoo
Would consider another tattoo
No plans to get/would not consider another tattoo
19 (16)
6 (32)
5 (26)
5 (26)
0
9 (47)
4 (21)
15 (83)
6 (32)
7 (37)
10 (53)
10 (53)
7 (37)
6 (32)
4 (21)
5 (26)
0
11 (58)
7 (37)
1 (5)
99 (84)
9 (9)
14 (14)
13 (13)
8 (8)
20 (20)
6 (6)
41 (43)
6 (6)
85 (86)
85 (86)
87 (88)
9 (9)
13 (13)
16 (16)
4 (4)
6 (6)
29 (29)
35 (35)
35 (35)
.02
NS
NS
NS
.02
NS
x2.002
NS
\.001
.002
.001
.004
NS
NS
.007
NS
x2.01
NS
x2.01
NS, Not statistically significant.
*Included weeping (fluid discharge), crusting, bleeding, swelling, scarring, itching, tenderness, redness, and sun sensitivity. No one said
a problem lasted longer than, and only people said a problem (swelling in 1 and itching in both) occurred later than, 2 weeks after a tattoo
was done.
Table IIIB. Associations of body piercing
N (%)
Total, No. (%)
First BP before age 18 y*y
Ever allowed a BP to close due to medical
problems during the first 3 wkz
Ever allowed a BP to close due to medical
problems later than 3 wkz
Ever allowed a BP to close for nonmedical
reasons
Considered allowing a BP to close for
nonmedical reasons
Drunk at time of first BP
High at time of first BP
72 (100)
21 (30)
4 (6)
1 (0.2)
34 (48)
3 (4)
3 (4)
4 (5)
BP, Body piercing.
*Total number of BPs does not add up to 72 because of 2 absent
responses.
yThere was no difference in any of the measures between those
who obtained their first BP younger than 18 years and those who
obtained it at or after 18 years of age.
zIncluded discomfort, pain, tenderness, redness, swelling, itching,
bleeding, weeping (fluid discharge,) and pus.
J AM ACAD DERMATOL
Laumann and Derick5
ARTICLE IN PRESS