Content uploaded by Christopher Lynn
Author content
All content in this area was uploaded by Christopher Lynn on May 06, 2019
Content may be subject to copyright.
Original Research Article
Tattooing to “Toughen Up”: Tattoo Experience and Secretory Immunoglobulin A
CHRISTOPHER D. LYNN,* JOHNNA T. DOMINGUEZ, AND JASON A. DECARO
Department of Anthropology, University of Alabama, Tuscaloosa, Alabama 35487
Objectives: A costly signaling model suggests tattooing inoculates the immune system to heightened vigilance
against stressors associated with soft tissue damage. We sought to investigate this “inoculation hypothesis” of tattooing
as a costly honest signal of fitness. We hypothesized that the immune system habituates to the tattooing stressor in
repeatedly tattooed individuals and that immune response to the stress of the tattooing process would correlate with
lifetime tattoo experience.
Methods: Participants were 24 women and 5 men (aged 18–47). We measured immune function using secretory
immunoglobulin A (SIgA) and cortisol (sCORT) in saliva collected before and after tattoo sessions. We measured tattoo
experience as a sum of number of tattoos, lifetime hours tattooed, years since first tattoo, percent of body covered, and
number of tattoo sessions. We predicted an inverse relationship between SIgA and sCORT and less SIgA immunosup-
pression among those with more tattoo experience. We used hierarchical multiple regression to test for a main effect of
tattoo experience on post-tattoo SIgA, controlling for pretest SIgA, tattoo session duration, body mass, and the interac-
tion between tattoo experience and test session duration.
Results: The regression model was significant (P50.006) with a large effect size (r
2
50.711) and significant and pos-
itive main (P50.03) and interaction effects (P50.014).
Conclusions: Our data suggest that the body habituates over time to the tattooing stressor. It is possible that indi-
viduals with healthy immune systems heal faster, making them more likely to get multiple tattoos. Am. J. Hum. Biol.
00:000–000, 2016. V
C2016 Wiley Periodicals, Inc.
Despite growing popularity of tattooing among all social
classes since the 1970s (DeMello, 2000; Hill, 1972; Rubin,
1988), biological studies of tattooing have been restricted
largely to health risks. Yet, historic and ethnographic
accounts have long associated some forms of tattooing
with “hardening” or protection against sickness (DeMello,
2000; Ludvico and Kurland, 1995). In fact, because of the
health risks involved, such as skin cellulitis, bacterial
infection, blood-borne disease transmission, hepatitis,
allergic reactions to carcinogenic colors, and hazardous
pigment concentrations (Kluger and Koljonen, 2012;
Kluger et al., 2012; Laux et al., 2015; Wohlrab et al.,
2009), successful tattooing might actually indicate resist-
ance to such dangers (Koziel et al., 2010).
Tattooing may stimulate the immune system in a man-
ner similar to a vaccination to be less susceptible to future
pathogenic infiltration. Tattoo aficionados widely report
becoming “addicted” to tattooing (e.g., http://www.newloo-
khouston.com/blog/2010/06/28/15-reasons-someone-could-
become-addicted-to-tattoos/), which we believe occurs pri-
marily among those whose tattoos remain attractive
because they heal quickly and cleanly. Tattoos may not lit-
erally harden or protect a body, but they may signal
underlying immunological and genetic quality (Koziel
et al., 2010). We tested this inoculation hypothesis in a
sample in the U.S. South using a pre-post tattoo session
design. Secretory immunoglobulin A (SIgA) and cortisol
(sCORT) in saliva were collected and compared to lifetime
tattoo experience. We sought to determine if tattooing
could play a role as a costly honest signal of quality by
influencing immunosuppression in response to the stress
of tattooing.
Costly honest signaling
Tattoos are a form of body ornamentation that has been
used throughout human history (Gilbert, 2000). Wester-
marck ([1891] 2003) suggested that such human self-
adornment serves as means of attraction, as with the
bright colors of birds and other sexually selected charac-
teristics in nonhuman animals. But tattooing is not
merely decorative; it is communicative, painful, and dan-
gerous. Costly signaling is one of the most salient themes
found in studies of body ornamentation (Carmen et al.,
2012). Signaling refers to communication between indi-
viduals whose interests may be in conflict, such as
between potential mates or predator and prey. Costly sig-
naling occurs when the production of the signal entails
some handicap to the signaler (Zahavi and Zahavi, 1997).
The health risks associated with tattoos are examples of
such costs (Laux et al., 2015). Death has even been
reported as a result of receiving tattoos as part of tradi-
tional, extensive, and temporally intensive rites. For
example, a Samoan pe’a is a tattoo for males covering the
entire lower torso and thighs that is administered by
hand tools. The continual stress of this administration,
which was once the only way to tattoo, has been reported
to weaken the abdominal walls and cause umbilical her-
nia (Moe, 1989). The ability to withstand such painful
forms of tattooing was often linked to ritual change in sta-
tus (Stirn, 2003).
Costly honest signals are those produced with high
degrees of fidelity because they are particularly hard or
impossible to fake (Zahavi and Zahavi, 1997). Tattoos that
heal with no signs of deleterious reaction are considered
*Correspondence to: Christopher D. Lynn, Department of Anthropology,
University of Alabama, Box 870210, Tuscaloosa, AL 35487, USA.
E-mail: cdlynn@ua.edu
Received 24 June 2015; Revision received 23 December 2015; Accepted
12 February 2016
DOI: 10.1002/ajhb.22847
Published online 00 Month 2016 in Wiley Online Library
(wileyonlinelibrary.com).
V
C2016 Wiley Periodicals, Inc.
AMERICAN JOURNAL OF HUMAN BIOLOGY 00:00–00 (2016)
beautiful and tattoo care websites commonly discuss how
to care for new tattoos so they heal properly without los-
ing color and clarity (e.g., http://www.skin-artists.com/tat-
too.htm). But, tattoos are wounds and must be cared for
like any wound to heal properly. Small lesions heal rela-
tively quickly, but deeper or more extensive surface inju-
ries take longer to heal. This time puts individuals at
greater risk of infection and scarring (Galili, 2015). Thus,
quick recovery from a tattoo may be a costly honest signal
of immunological health and phenotypic vigor, especially
when the tattoo is large or administration is intensive.
Several studies of this costly signaling model of tattoo-
ing have been conducted with mixed results. Ludvico and
Kurland (1995) found some ethnologic relationships
between sexual selection and scarification, which included
tattooing as one form. In a study rating phenotypically
male and female computer-simulations with and without
tattoos (Wohlrab et al., 2009), tattooed men were consid-
ered healthier than nontattooed men by female raters and
more dominant than nontattooed men by both sexes. In a
study correlating tattooing and piercing with biological
quality, Koziel et al. (2010) found tattooing but not pierc-
ing significantly and positively associated with greater
body symmetry, an indication of developmental stability
and health that is highly correlated with attractiveness.
Immunosuppression and stress
A more direct test of the costly signaling model would
involve assessing the effect of tattooing on the immune
system. Onset of stress, such as is involved in tattooing, is
followed within the first few minutes by enhancement of
immune function above baseline. Cortisol is released in
humans as part of the hypothalamic-pituitary-adrenal
stress response 30–60 min after stressor onset and func-
tions, in part, to suppress this immune response and
restore it to baseline (Sapolsky, 2002). Studies of SIgA and
sCORT responses to stress indicate that they display
inverse patterns before and after stress response (Ng
et al., 2003; Watanuki and Kim, 2005). However, this
inverse association is only statistically significant in
awakening diurnal levels (Hucklebridge et al., 1998). This
lack of correlation is likely due to their asynchronous
onset/offset systems—unlike the latency of cortisol pro-
duction, SIgA is immediately responsive.
Immunoglobulin A is a polymeric antibody produced in
bodily mucosa and serum and the frontline defense of the
upper respiratory and gastrointestinal tracts (Trochimiak
and H€
ubner-Wozniak, 2012; Woof and Kerr, 2006). IgA
antibodies provide protection against a range of pathogens
and toxins through binding to immunoglobulin receptors
on the basolateral surfaces of epithelial cells of the mucosa
(Woof, 2013). SIgA elevation is generally associated with
acute, temporary stress (Bristow et al., 1997), with immu-
nosuppression occurring 30–60 min after stress onset.
This is followed by SIgA elevation again above baseline
(Trochimiak and H€
ubner-Wozniak, 2012) when cortisol
production ceases (again, with the same 30–60 min latency
after stressor offset). The degree of immunosuppression
during stress response is, importantly, dependent on a
number of factors, such as valence of and habituation to
the stressor (Beck et al., 2000; Gleeson, 2007).
Prolonged stress or physical activity exposure and
stress associated with negative mood have been repeat-
edly associated with prolonged immunosuppression and
low SIgA levels (Bristow et al., 1997; Trochimiak and
H€
ubner-Wozniak, 2012). Studies of overtraining in the
military (Carins and Booth, 2002) and athletics (Spence
et al., 2007) support this, as they find extended physical
strain associated with upper respiratory tract infections
(URTIs—e.g., common colds) and immunosuppression in
SIgA. Tattooing presents a similar paradigm. Tattooists
consulted for this study indicate the tattoo process can
produce a feeling of being “wiped out.” Symptoms akin to
URTIs are anecdotally reported in conjunction with get-
ting new tattoos, especially extensive tattoos taking long
periods of time or covering much of the body. Given that
tattooing sessions last long enough for the immunosup-
pressive phase of the stress response to be engaged by the
end of the session, we would generally expect people
undergoing the acute stress of tattooing to show signs of
immunosuppression immediately upon completion of a
tattooing session. But, as research demonstrates that
exposure to repeated stressors can result in immune
response habituation (Gleeson, 2007; McEwen, 2004), we
predict less immunosuppression among individuals with
more tattoo experience.
This can be tested by comparing SIgA and sCORT levels
before and after tattooing in individuals with varying tat-
too experience. Both biomarkers can be sampled easily
from saliva (Kugler et al., 1992). Saliva can provide a gen-
eral reflection of the entire mucosal immune system
because salivary glands in the mouth are dense with
immunoglobulin A-producing plasma cells and ductal and
acinar cells with high levels of IgA receptors (Mestecky,
1993). SIgA is flow-dependent, meaning that the rate of
salivation is important in measuring its quantity, but flow
rate can be controlled for by timing saliva flow or meas-
uring IgA against the quantity of other elements of the
saliva sample, such as total protein (Brandtzaeg, 1971).
Immune response has also been associated with body
mass (Nazmi and Victora, 2007), age-related factors
(Blackwell et al., 2010), and baseline immunological
health (Koziel et al., 2010; Manning, 2002; Palmer and
Strobeck, 1986; Schaap et al., 2006), so these must be
accounted for.
As part of a larger study of cultural models held about
tattooing, we sampled saliva among participants immedi-
ately before and after tattooing sessions to test for the
degree of immunosuppression related to the tattoo. We
predicted that people with more extensive previous tattoo-
ing experience would display the habituation or inocula-
tion effect and suffer less immunosuppression as a result
of the tattoo stress.
METHODS
From May through December 2012, we collected data
from three tattoo studios in Leeds and Tuscaloosa, Ala-
bama that had granted permission to collect data on their
premises. Since the larger study was focused on cultural
models about tattooing among females in the U.S. South-
east, the majority of the sample comprises women. The
University of Alabama Institutional Review Board
approved research protocols.
Participants and recruitment
We used snowball sampling to recruit 24 women and 5
men (aged 18–47) receiving tattoos at the three studios.
We used social media (Facebook and Twitter) to find
2C.D. LYNN ET AL.
American Journal of Human Biology
participants who would be receiving tattoos at the Tusca-
loosa studios and made arrangements to be on site to col-
lect data when they received the tattoo. The owner of the
Leeds studio alerted us when clients were scheduled and
gave permission to recruit on site. Three participants
received multiple tattoos (2–3) during the study and con-
tributed data each time. Each of those participants’ data
were averaged to provide one statistic per variable. Par-
ticipation was voluntary, and all participants gave
informed consent.
Tattoo experience and saliva collection
We collected demographic, tattoo experience, and
anthropometric data in the tattoo studios before each par-
ticipant’s tattoo session. Using paper and pencil surveys,
we queried number of tattoos, number of tattoo sessions,
lifetime hours spent receiving tattoos, years since first tat-
too, and percent of body tattooed. These values were
summed to create a tattoo experience variable. We col-
lected saliva samples immediately before and after each
session using commercially available SalivaBio inert poly-
mer oral swabs and storage tubes (Salimetrics LLC, State
College, PA). Following manufacturer recommendations
(Salimetrics and SalivaBio, 2015), participants were
asked to place the swab under their tongues for 1–2 min
without chewing to ensure saturation. They then placed
the swab in the basket insert in the upper portion of the
SalivaBio storage tube. We recorded the times of the pre-
and post-tattoo measures to control for the length of time
of the tattoo session, duration between biomarker meas-
ures, and diurnal patterns of SIgA and cortisol. We stored
tubes in an insulated bag with a frozen ice pack while at
the field sites, and then transferred them to 2308C freez-
ers at the University of Alabama, where they were stored
until analysis.
Large tattoos often involve longer and multiple sessions.
Therefore, people with more tattoo experience may also sit
for longer sessions, creating an interaction between tattoo
experience and session duration. We tested this by creating
interaction terms for inclusion in analysis.
Immunosuppression analysis
We measured immunosuppression using posttest SIgA
while controlling for pretest SIgA. Based on the principles
of immunosuppression during stress outlined above, we
predicted SIgA levels would be lower after the tattoo ses-
sion (posttest) than before it (pretest) in people with lim-
ited tattoo experience and neutral or elevated in people
with more tattoo experience, who would be habituated to
such stress. We controlled for flow rate by measuring the
total amount of protein in the saliva following Brandtzaeg
(1971), calculating the SIgA pretest measure as SIgA
pret-
est
/Protein
pretest
, the posttest measure as SIgA
posttest
/Pro-
tein
posttest
, and mean SIgA change as:
SIgAposttest
Proteinposttest
2SIgApretest
Proteinpretest
SIgApretest
Proteinpretest
Cortisol analysis was not part of the original study
design, but, because of its direct effect as an immunosup-
pressant, we assayed it to obtain a picture of the interac-
tion among the tattoo and stressor, physiological stress,
and SIgA response. However, there were not sufficient
salivary volumes in all remaining samples. Thus, we
obtained sCORT sufficient to calculate pre-posttest differ-
ence in 22 participants. In assessing the influence of corti-
sol on immune response, we predicted sCORT would
increase pre-posttest and have an inverse pre-posttest
relationship with SIgA in participants with lower lifetime
tattoo experience.
Anthropometrics and potential covariates
We collected information about demographics and other
factors that might influence immune and stress response,
such as alcohol, tobacco, marijuana, or medication use;
recent illness; and perceived stress (Cohen and William-
son, 1988). Demographic information included age, eth-
nicity, relationship status and appraisal, education, and
social status (Singh-Manoux et al., 2003). We collected
body mass, handgrip strength, digit ratios, and bilateral
symmetry data to control for baseline immune response
(Innes, 1999; Koziel et al., 2010; Manning, 2002; Nazmi
and Victora, 2007; Palmer and Strobeck, 1986; Schaap
et al., 2006). We measured participant weight in light
clothing (without shoes) using a Tanita Model TBF 310
portable bioimpedance analyzer that calculated body
mass index (BMI, kg/m
2
) using self-reported height. We
measured handgrip strength with a Detecto Model DHS
174 portable hand dynamometer (kg). The results of two
trials on each hand were averaged together to determine
a rating. We measured 2nd-to-4th digit ratios (2D4D) and
bilateral symmetry by scanning participant hands using a
Canon CanoScan LiDE 700F flatbed scanner. We used the
measuring tool in Adobe Acrobat Pro Version 11.0.11 to
calculate the length of the 2nd and 4th finger of each
hand from tip to basal crease. We calculated 2D4D as
P2D=4D
ðÞ
i
2;where i5respective hand:We calculated fluc-
tuating asymmetry (FA or deviation from symmetry) fol-
lowing Palmer and Strobeck (1986), as PRi-Li
ðÞ
Number of pairs,
where i5respective finger. Unfortunately, handscans of
six participants were not properly saved and 2D4D and
FA could not be calculated for them.
Biomarker assays
We assayed the samples with commercially available
SIgA, cortisol, and total protein kits (Salimetrics LLC,
State College, PA and Pierce Biotechnology, Rockford, IL).
Prior to analysis, we centrifuged the samples at 3,000 rpm
for 15 min to remove mucins, and all samples were assayed
in duplicate. We pipetted 25 mL of saliva into 96-well micro-
titer plates precoated with highly purified human SIgA,
followed by a goat anti-human SIgA antibody conjugated to
horseradish peroxidase. We washed the wells and detected
free antihuman SIgA by adding tetramethylbenzidine to
each well, stopping the reaction following incubation using
a sulfuric acid solution, and determining optical density
(450 nm) via a PowerWave HT Microplate Spectrophotome-
ter (BioTek Instruments, Winooski, VT). All standards,
controls, and unknowns were run in duplicate, and out-
comes represent the averages. We used wells containing
known high and low SIgA concentrations to correct for
multiple plate comparisons. Intra- and inter-assay coeffi-
cients of variation (CVs) were below 12%.
We assayed salivas for cortisol using the Salimetrics
Expanded Range High Sensitivity Salivary Cortisol
TATTOO EXPERIENCE AND SECRETORY IMMUNOGLOBULIN A 3
American Journal of Human Biology
enzyme immunoassay (cat #1-3002). This is a competitive
immunoassay in which cortisol in the sample competes for
anti-cortisol antibody binding sites with a known quantity
of horseradish perodixase-linked cortisol. The horseradish
peroxidase substrate tetramethylbenzidine develops color,
measured at 450 nm, in inverse proportion to the quantity
of cortisol in the original sample. We used a BioTek
Powerwave HT microplate reader for absorbance detec-
tion. Intra-assay CVs were below 10%, and inter-assay
CVs were below 13%.
We assayed for total protein using the Pierce BCA pro-
tein assay kit. This assay relies on the reduction of copper
by protein that occurs in an alkaline solution. The quan-
tity of protein, directly proportional to the reduction of
Cu
12
to Cu
11
, is calculated against an albumin standard
curve by quantifying Cu
11
through a timed reaction with
bicinchoninic acid that generates an absorbance peak at
562 nm.
Analysis
We calculated descriptive statistics for all variables to
characterize the sample. Data for subjects who contrib-
uted multiple times were tested independently and as
averages, and there were no significant differences in
analyses. Therefore, averages for these three participants’
multiple tattoo sessions were retained for analysis. We
conducted tests for normality, linearity, and homoscedas-
ticity to ensure the underlying assumptions of multivari-
able analysis were met and transformed non-normal
variables using log
10
after adding 1 to ensure constancy in
valence. Transformed variables included tattoo experience
measures, session duration, and sCORT. We used bivari-
ate correlations to compare sCORT and SIgA to tattoo
experience and test for predicted inverse relationships. To
test the hypothesis that tattoo experience is associated
with immunosuppression, we conducted hierarchical mul-
tiple regression on posttest SIgA. The first block included
pretest SIgA and any demographic or anthropometric
covariates that fit the model. Other covariates were cho-
sen using stepwise methods. Block 2 included tattoo expe-
rience and session duration. Block 3 included the tattoo
experience-by-session duration interaction term. All vari-
ables were standardized using Z-scores. Interaction term
was calculated as the cross-product of the standardized
Block 2 variables. All analyses were conducted using
SPSS Version 21 (IBM Corp., 2012) and a50.05.
RESULTS
The sample was mostly white, young, educated, in a
committed relationship, and middle-class. Mean age
(6SD) was 26.38 67.15. Twenty-seven participants were
white (84%), and 17 (64%) had at least some college. Par-
ticipants averaged 5.33 61.66 on a 10-rung scale in self-
reported social status. Forty-one percent were married or
in a committed relationship. These demographic data did
not significantly correlate with any SIgA variables and
were not used in subsequent analysis. Participants
ranged widely in self-reported tattoo experience (Table 1).
Several were getting their first tattoo (38%), while one
had 240 h under the needle.
We used bivariate correlations (two-tailed) to test for
significant influences of medication, alcohol, tobacco, or
other drug use on SIgA, sCORT, and perceived stress; to
test relationships between sCORT and SIgA; and to
choose regression model variables. There were significant
relationships between pre-tattoo sCORT and alcohol in
the last 24 h (r50.67, P<0.001) and post-tattoo sCORT
and alcohol per week (r50.42, P50.04) and in the last
24 h (r50.51, P50.01). Since alcohol and cigarette use
were not correlated with any SIgA measures in bivariate
correlations or model testing, they were not used in SIgA
analyses. Although we imagine the anticipation of getting
a tattoo would influence pretest cortisol, there were no
other significant relationships with pretest sCORT. There
were no significant relationships between pre-posttest dif-
ferences in sCORT and SIgA (r520.23, P50.30) (Fig. 1).
Comparison of SIgA and sCORT and tattoo experience
variables (Table 2) indicates significant positive correla-
tions between posttest SIgA and percent of body covered
(P50.02) and significant inverse correlations between
SIgA change and number of sessions tattooed (P50.003),
hours tattooed (P50.002), percent of body tattooed
TABLE 1. Parameters of untransformed study variables
Mean 6SD Min Max
Tattoo
experience
Years 4.26 66.27 0 26
Number 2.76 63.07 0 10
Sessions 4.92 611.91 0 60
Hours 20.08 657.70 0 240
Percent
body
covered
0.17 60.57 0 2
Session duration (hours) 1.49 61.15 0.09 4.04
SIgA Pretest 0.016 60.007 0.00 0.03
Posttest 0.017 60.008 0.00 0.04
sCORT Pretest 0.106 60.102 0.004 0.404
Posttest 0.140 60.133 0.003 0.428
BMI (kg/m
2
) 25.47 66.47 16.20 39.67
Handgrip strength (kg) 61.88 617.36 39.83 108.45
Digit ratio (2D4D) 0.977 60.03 0.92 1.03
Fluctuating asymmetr y 0.010 600.06 20.10 0.10
Perceived stress 4.43 63.56 0 14
SIgA values expressed in relation to total protein.
Fig. 1. Pretest, posttest, and difference levels of sCORT (log
10
) and
SIgA (proportion of protein). Note that SIgA change is not calculated
directly as pre-posttest but as proportion of whole protein. Differen-
ces are not significant.
4C.D. LYNN ET AL.
American Journal of Human Biology
(P50.002), and total tattoo experience (P50.04). Body
density, handgrip strength, and FA were not associated
with SIgA, but 2D4D was significantly and negatively
associated with pretest SIgA (r520.475, P50.04) and
SIgA change (r520.626, P50.004).
We used hierarchical multiple regression to test the
influence of tattoo experience on posttest SIgA. As indi-
cated in Table 3, we included pretest SIgA as a control in
Block 1 and used stepwise methods to select from among
potential demographic and anthropometric covariates.
BMI provided the best fit and is also included in Block 1.
Neither predictor nor the Block 1 model were significant
(F
2,15
52.548, P50.112). Block 2 includes tattoo experi-
ence and session duration. The Block 2 model
(F
4,13
53.381, P50.042) was significant, as were the vari-
ables pretest SIgA and tattoo experience. Block 3 included
the tattoo experience-by-session duration interaction
term; the model (F
5,12
55.909, P50.006), pretest SIgA,
BMI, tattoo experience, and the interaction were signifi-
cant predictors in block 3.
To examine the nature of the interaction effect, we plot-
ted tattoo experience and tattoo experience-by-session
duration at 61 SD (http://www.jeremydawson.co.uk/
slopes.htm). As Figure 2 illustrates, there is a greater ele-
vation in posttest SIgA among those with more tattoo
experience during longer tattoo session.
Because our sample size for 2D4D and sCORT was
smaller, we used stepwise methods to test their influences
on posttest SIgA, using Bonferroni correction for multiple
analyses (a50.025). We tested 2D4D, pretest sCORT,
posttest sCORT, and sCORT change in models including
pretest SIgA, tattoo experience, session duration, and tat-
too experience-by-session duration. Only the model
including digit ratio was significant (F
5,9
54.576,
P50.024, r
2
50.718). The only variable that significantly
predicted posttest SIgA in these models was tattoo experi-
ence (b50.531, P50.025) in the model that included
sCORT change (F
5,9
54.386, P50.027, r
2
50.709).
DISCUSSION
We tested the hypothesis that there would be less
immunosuppression among those with more tattoo experi-
ence. Tattoo experience correlated positively with items
comprising post-tattoo experience but no pre-tattoo meas-
ures. Although there was a nonsignificant decrease from
pre-posttest in SIgA, there was a significant positive cor-
relation between tattoo experience and posttest SIgA
when controlling for the pretest measure. The sample size
was small, but these data are important in providing into
the body’s physiological response to tattooing. There are a
few ways to interpret these findings. One possibility is
that SIgA is generally downregulated but more specifi-
cally responsive in older participants, who were generally
those with the most tattoo experience and whose tattoo
sessions were longer in our study. Studies show that life
history factors influence trade-offs in immune response
and that energy allocation for SIgA production may be
diminished in older participants because of reduction in
age-related innate immune function. For instance,
TABLE 3. Hierarchical multiple regression of tattoo experience on
posttest SIgA, controlling for pretest SIgA, tattoo session duration (h),
BMI, and tattoo experience-by-session duration interaction
Standardized bPr
2
Block 1 0.254
SIgA
pretest
0.452 0.080
BMI 0.451 0.081
Block 2 0.510
SIgA
pretest
0.493 0.036
BMI 0.330 0.156
Tattoo experience 0.445 0.050
Tattoo session duration 0.285 0.166
Block 3 0.711
SIgA
pretest
0.459 0.018
BMI 0.385 0.049
Tattoo experience 0.406 0.030
Tattoo session duration 0.206 0.215
Tattoo experience-
x-session duration
0.461 0.014
Note: All variables transformed using log
10
.
Fig. 2. Tattoo experience (unstandardized b50.380) and session
duration (0.192) plotted at 61 SD to visualize interaction effect
(0.359). Variables are transformed (log
10
) and standardized. (http://
www.jeremydawson.co.uk/slopes.htm).
TABLE 2. Bivariate correlations between log
10
-transformed SIgA, sCORT, and tattoo experience variables
Tattoo experience
Session
durationYears Number Sessions Hours
Percent
body covered Total
Pretest 20.283 20.239 20.280 20.263 20.148 20.279 20.017
SIgA Posttest 0.154 0.229 0.402 0.430 0.534
b
0.356 0.359
Change 20.379 20.419 20.624
a
20.638
a
20.650
a
20.577
a
20.260
sCORT Pretest 20.286 20.316 20.297 20.274 20.095 20.301 20.257
Posttest 20.189 20.171 20.228 20.188 20.230 20.206 20.288
Change 20.194 20.224 20.044 20.026 0.360 20.084 0.180
a
P<0.01.
b
P<0.05 (two-tailed).
TATTOO EXPERIENCE AND SECRETORY IMMUNOGLOBULIN A 5
American Journal of Human Biology
Blackwell et al. (2010) have found greater negative effects
for older ages in immunoglobulin E response. Immuno-
globulin E is associated mainly with allergic responses
and, like IgA, is found extensively in mucous membranes.
However, the relationships between age and SIgA meas-
ures and age and tattoo experience in our study were not
statistically significant, supporting the interpretation
that it is the tattoo experience that is important.
Thus, the second interpretation of these data is in light of
the inoculation model of tattooing. We predicted a greater
influence on immune response among those with more tat-
too experience, which was supported. The effect was greater
when the duration of the tattoo session was longer. Tattoo
collectors tend to get larger tattoos that take longer to
administer, often over multiple sessions. The immunologi-
cal boost associated with this interaction between tattoo
experience and session duration has been born out in ani-
mal studies. Administration of vaccinations via the same
technique used in tattooing to inject ink under the skin is a
more effective method of vaccination than intramuscular
injection. Rather than restrict the inoculation effect to the
specific agent being injected, tattooing transfects more cells
due to its larger application area and produces a more gen-
eralized immune response (Pokorna et al., 2008; van den
Berg et al., 2014). Our data may confirm the numerous his-
torical and cultural beliefs associating tattoos with protect-
ing the body, rather than injuring it (DeMello, 2000;
Ludvico and Kurland, 1995). Accounts dating to the 19th
century report of young people skirting legality to get tat-
tooed for purposes of “toughening up” (e.g., Parry, [1933]
2006; Smeaton, 1898; Steward, 1990; Vale and Juno, 1989).
Among military personnel and others who value toughness
for their safety or livelihoods, tattoos have represented the
ability to withstand sickness or disease and to protect
against and recover from illness (Parry, [1933] 2006).
Tattooing is not unique in this protective effect, as simi-
lar relationships have been observed between SIgA and
exercise (Bishop and Gleeson, 2009; Gleeson and Pyne,
2000; Leite et al., 2013) and choral singing (Beck et al.,
2000). Among elite athletes, postexercise IgA suppression
is associated only with prolonged bouts (>1.5 h) and low
food intake (Gleeson, 2007), whereas, among highly
trained choral singers, positive stress was more likely to
lead to elevated SIgA than negative stress and extended
relaxation practices were more associated with higher
SIgA than shorter term practices (Beck et al., 2000).
Based on those studies, we can consider our data in two
ways. First, participants with greater tattoo experience
may be more excited than anxious about a tattooing ses-
sion, resulting in reduced immunosuppression. Another
explanation, which is not mutually exclusive, is that peo-
ple with higher tattoo experience might also display
reduced IgA suppression after tattooing, similar to elite
athletes who habituate to moderate and high intensity
exercise stress over time (Gleeson, 2000).
The relationship among tattooing, immune response,
and athletics is no coincidence. Competition and tattooing
are ways to demonstrate fitness, and tattoos may amplify
the fitness signal. For instance, Mayers et al. (2002) found
male athletes significantly more likely than male nonath-
letes to be tattooed. In contemporary North America and
Europe, healthcare innovations have changed the grain
and visibility of disparities (Sridhar, 2005) in a way that
potentially lessens the salience of subtle signals of biologi-
cal quality. Thus, tattooing may “up the ante,” as Carmen
et al. (2012) suggest, by drawing attention to one’s capacity
to undergo hardship and heal. Although we did not test the
tendency of individuals who heal well to be more likely to
collect tattoos, tattoo artists consulted for this study indi-
cate that tattoo collectors do tend to heal quickly from tat-
toos. Well-done tattoos that heal cleanly draw compliments
from others, which contributes to positive feedback, lead-
ing them to get even more tattoos. In contrast, tattooing is
an injury to the skin that can become infected or stimulate
immune responses that damage the site and thus the ulti-
mate appearance of tattoo (e.g., keloids or granulomas)
(Goldstein, 1979; LeBlanc et al., 2012).
There are several limitations of this study, leading us to
be cautious lest we over-interpret these findings. The
sample was exclusively young, white, educated, middle-
class, and predominantly female tattoo novices, so follow-
up studies should include more varied demographic com-
parisons and people with more extensive tattoo experi-
ence. The region from which participants were recruited
involved only cities in Alabama, where tattooing hygiene
standards are high. Future research should examine tat-
tooing culture in regions where tattoo infections are more
common due to poor tattoo sanitation and with longstand-
ing traditions of extensive tattooing. Finally, SIgA may be
sensitive to dietary and diurnal influences (Gleeson, 2000)
that field conditions made it difficult to control for but
which should be accommodated where possible. Yet,
through these data, we can better understand the rela-
tionship between tattooing and stress and why tattooing
could be a reliable signal of quality. Tattooing takes a toll
on the body, so more fit individuals may have immune sys-
tems better able to adapt to the additional strain. We
anticipate future research will validate this immunologi-
cal importance of tattooing as embodied culture.
ACKNOWLEDGMENTS
The authors would like to thank Cynical, Inkheart, and
Showcase Tattoos for permission to recruit among their
clients for this project and for allowing us to collect data
on site. The authors would also like to thank students in
the Human Behavioral Ecology Research Group for assis-
tance in research design and data collection and two
anonymous reviewers for helpful critique of a previous
draft of this article. Previous analyses of these data were
presented at the 2015 meeting of the Human Biology
Association. The authors are grateful for helpful com-
ments from session attendees on that presentation. The
authors have no conflicts of interest with regard to any
entities involved in this study.
AUTHOR CONTRIBUTIONS
CDL designed the study and directed the implementa-
tion and data collection. JTD collected the data. JAD con-
ducted the biochemical analysis, edited the manuscript,
and provided critical feedback. CDL and JTD analyzed
the data and drafted the manuscript.
LITERATURE CITED
Beck RJ, Cesario TC, Yousefi A, Enamoto H. 2000. Choral singing, per-
formance perception, and immune system changes in salivary immuno-
globulin A and cortisol. Music Percept Interdiscip J 18:87–106.
Bishop NC, Gleeson M. 2009. Acute and chronic effects of exercise on
markers of mucosal immunity. Front Biosci 14:4444–4456.
6C.D. LYNN ET AL.
American Journal of Human Biology
Blackwell AD, Snodgrass JJ, Madimenos FC, Sugiyama LS. 2010. Life his-
tory, immune function, and intestinal helminths: trade-offs among
immunoglobulin E, C-reactive protein, and growth in an Amazonian
population. Am J Hum Biol 22:836–848.
Brandtzaeg P. 1971. Human secretory immunoglobulins-VII: concentra-
tions of parotid IgA and other secretory proteins in relation to the rate of
flow and duration of secretory stimulus. Arch Oral Biol 16:129521310.
Bristow M, Hucklebridge F, Clow A, Evans PD. 1997. Modulation of secre-
tory immunoglobin A in saliva in relation to an acute episode of stress
and arousal. J Psychophysiol 11:248–255.
Carins J, Booth C. 2002. Salivary immunoglobulin-A as a marker of stress
during strenuous physical training. Aviat Space Environ Med 73:1203–
1207.
Carmen RA, Guitar AE, Dillon HM. 2012. Ultimate answers to proximate
questions: the evolutionary motivations behind tattoos and body pierc-
ings in popular culture. Rev Gen Psychol 16:134–143.
Cohen S, Williamson GM. 1988. Perceived stress in a probability sample of
the United States. In: Spacapan S, Oskamp S, editors. The social psy-
chology of health. Newbury Park, CA: Sage.
DeMello M. 2000. Bodies of inscription: a cultural history of the modern
tattoo community. Durham, NC: Duke University Press.
Galili U. 2015. Acceleration of wound healing by-gal nanoparticles inter-
acting with the natural anti-gal antibody. J Immunol Res 2015:Article
ID 589648. doi: 10.1155/2015/589648 [Epub ahead of print].
Gilbert S. 2000. Tattoo history: a source book: an anthology of historical
records of tattooing throughout the world. New York; Berkeley, CA:
Juno Books.
Gleeson M. 2000. Mucosal immune responses and risk of respiratory ill-
ness in elite athletes. Exer Immunol Rev 6:5–42.
Gleeson M. 2007. Immune function in sport and exercise. J Appl Physiol
103:693–699.
Gleeson M, Pyne DB. 2000. Special feature for the olympics: effects of exer-
cise on the immune system: exercise effects on mucosal immunity.
Immunol Cell Biol 78:536–544.
Goldstein N. 1979. IV. Complications from tattoos. J Dermatol Surge Oncol
5:869–878.
Hill A. 1972. Tattoo renaissance. In: Lewis G, editor. Side-saddle on the
gold calf. Pacific Palisades, CA: Goodyear.
Hucklebridge F, Clow A, Evans P. 1998. The relationship between salivary
secretory immunoglobulin A and cortisol: neuroendocrine response to
awakening and the diurnal cycle. Int J Psychophysiol 31:69–76.
Innes E. 1999. Handgrip strength testing: A review of the literature. Aust
Occup Ther J 46:120–140.
Kluger N, Hubiche T, Del Giudice P. 2012. Tattoo-induced edema of the
lower limbs mimicking cellulitis: report of two cases. Int J Dermatol 52:
384–386.
Kluger N, Koljonen V. 2012. Tattoos, inks, and cancer. Lancet Oncol 13:
e161–e168.
Koziel S, Kretschmer W, Pawlowski B. 2010. Tattoo and piercing as signals
of biological quality. Evol Hum Behav 31:187–192.
Kugler J, Hess M, Haake D. 1992. Secretion of salivary immunoglobulin A
in relation to age, saliva flow, mood states, secretion of albumin, cortisol,
and catecholamines in saliva. J Clin Immunol 12:45–49.
Laux P, Tralau T, Tentschert J, Blume A, Al Dahouk S, B €
aumler W,
Bernstein E, Bocca B, Alimonti A, Colebrook H. 2015. A medical-
toxicological view of tattooing. Lancet 387:395–402.
LeBlanc PM, Hollinger KA, Klontz KC. 2012. Tattoo ink–related infec-
tions—awareness, diagnosis, reporting, and prevention. New Engl J
Med 367:985–987.
Leite MF, Aznar LCA, Ferreira MCD, Guar
e RO, Santos MTB. 2013.
Increased salivary immunoglobulin A and reduced a-amylase activity in
whole saliva from spastic cerebral palsy individuals. J Oral Pathol Med
42:480–485.
Ludvico LR, Kurland JA. 1995. Symbolic or not-so-symbolic wounds: the
behavioral ecology of human scarification. Ethol Sociobiol 16:155–172.
Manning JT. 2002. Digit ratio: a pointer to fertility, behavior, and health.
New Brunswick, NJ: Rutgers University Press.
Mayers LB, Judelson DA, Moriarty BW, Rundell KW. 2002. Prevalence of
body art (body piercing and tattooing) in university undergraduates and
incidence of medical complications. Mayo Clin Proc 77:29–34.
McEwen BS. 2004. Protective and damaging effects of the mediators of
stress and adaptation: allostasis and allostatic load. In: Schulkin J, edi-
tor. Allostasis, homeostasis, and the costs of physiological adaptation.
Cambridge: Cambridge University Press. p 65–98.
Mestecky J. 1993. Saliva as a manifestation of the common mucosal
immune system. Ann N Y Acad Sci 694:184–194.
Moe I. 1989. Samoan navel tattoo. In: Vale V, Juno A, editors. Modern
primitives: an investigation of contemporary adornment and ritual. San
Francisco, CA: RE/Search Publications. p 117–119.
Nazmi A, Victora CG. 2007. Socioeconomic and racial/ethnic differentials
of C-reactive protein levels: a systematic review of population-based
studies. BMC Public Health 7:212.
Ng V, Koh D, Mok BY, Chia SE, Lim LP. 2003. Salivary biomarkers associ-
ated with academic assessment stress among dental undergraduates.
J Dent Educ 67:1091–1094.
Palmer AR, Strobeck C. 1986. Fluctuating asymmetry: measurement,
analysis, patterns. Annu Rev Ecol Syst 17:391–421.
Parry A. [1933] 2006. Tattoo: secrets of a strange art. Mineola, NY: Dover.
Pokorna D, Rubio I, M€
uller M. 2008. DNA-vaccination via tattooing indu-
ces stronger humoral and cellular immune responses than intramuscu-
lar delivery supported by molecular adjuvants. Genet Vaccines Ther 6:
1–8.
Rubin A. 1988. The tattoo renaissance. In: Rubin A, editor. Marks
of civilization: artistic transformations of the human body. Los
Angeles, CA: University of California, Museum of Cultural History. p
233–262.
Salimetrics LLC, SalivaBio LLC. 2015. Saliva collection and handling
advice. Available at https://www.salimetrics.com/assets/documents/
Saliva_Collection_Handbook.pdf. Accessed on October 27, 2015.
Sapolsky RM. 2002. Endocrinology of the stress-response. In: Becker JB,
Breedlove SM, Crews D, McCarthy MM, editors. Behavioral endocrinol-
ogy. Cambridge, MA: MIT Press.
Schaap LA, Plujim SMF, Deeg DJH, Visser M. 2006. Inflammatory
markers and loss of muscle mass (sarcopenia) and strength. Am J Med
119:526.e9–526.e17.
Singh-Manoux A, Adler NE, Marmot MG. 2003. Subjective social status:
its determinants and its association with measures of ill-health in the
whitehall II study. Soc Sci Med 56:1321–1333.
Smeaton O. 1898. Tattooing and its history. Westminst Rev 149:320–323.
Spence L, Brown WJ, Pyne DB, Nissen MD, Sloots TP, McCormack JG,
Locke AS, Fricker PA. 2007. Incidence, etiology, and symptomatology of
upper respiratory illness in elite athletes. Med Sci Sports Exerc 39:577.
Sridhar D. 2005. Inequality in the United States Healthcare System. New
York: Human Development Report Office (HDRO), United Nations
Development Programme (UNDP).
Steward SM. 1990. Bad boys and tough tattoos: a social history of the tat-
too with gangs, sailors, and street-corner punks 1950-1965. Bingham-
ton, NY: Harrington Park Press.
Stirn A. 2003. Body piercing: medical consequences and psychological
motivations. Lancet 361:1205–1215.
Trochimiak T, H €
ubner-Wozniak E. 2012. Effect of exercise on the level of
immunoglobulin a in saliva. Biol Sport 29:255–261.
Vale V, Juno A. 1989. Lyle tuttle. In: Vale V, Juno A, editors. Modern primi-
tives: an investigation of contemporary adornment and ritual. San Fran-
cisco, CA: RE/Search. p 114–117.
van den Berg JH, Oosterhuis K, Schumacher TNM, Haanen JBAG, Bins
AD. 2014. Intradermal vaccination by DNA tattooing. Methods Mol Biol
1143:131–140.
Watanuki S, Kim Y. 2005. Physiological responses induced by pleasant
stimuli. J Physiol Anthropol Appl Hum Sci 24:135–138.
Westermarck E. [1891] 2003. History of human marriage. Whitefish, MT:
Kessinger.
Wohlrab S, Fink B, Kappeler PM, Brewer G. 2009. Perception of human
body modification. Pers Individ Dif 46:202–206.
Woof JM. 2013. Immunoglobulin A: molecular mechanisms of function and
role in immune defence. In: Nimmerjahn F, editor. Molecular and cellu-
lar mechanisms of antibody activity. New York: Springer. p 31–60.
Woof JM, Kerr MA. 2006. The function of immunoglobulin A in immunity.
J Pathol 208:270–282.
Zahavi A, Zahavi A. 1997. The handicap principle: a missing piece of Dar-
win’s puzzle. New York: Oxford University Press.
TATTOO EXPERIENCE AND SECRETORY IMMUNOGLOBULIN A 7
American Journal of Human Biology
