Tattooing and risk of hepatitis B: a systematic review and meta-analysis.
Jafari, Siavash ; Buxton, Jane A. ; Afshar, Kourosh 等
Tattooing and body arts have become more prevalent in recent years,
their popularity increasing among young adults. A population-based study
revealed that one third of people younger than 30 years old in the
United States have at least one tattoo. (1) Canadian data indicate that
around 8% of high school students have at least one tattoo (2,3) and
that 21% of those who did not have a tattoo were eager to have one.
Tattooing requires injection of pigments into the dermal layer of skin
by repeated puncture of the skin. Such close contact of the tattoo
instruments with blood and bodily fluids may cause transmission of viral
and bacterial infections if the instruments are used on more than one
person without being sterilized.
In North America, most cases of hepatitis B infection occur via
blood or sexual contact. (4) Acute hepatitis B infection causes chronic
infection in 6-10% of adolescents and adults; chronic infection can lead
to liver cirrhosis and to cancer. (5) Thus, cases of hepatitis B
infection due to tattooing have important clinical and public health
implications. Results from epidemiologic studies regarding the risk of
hepatitis among tattooed individuals are conflicting; (3,6) therefore,
we conducted a review of the current literature in order to quantify, in
a systematic fashion and by using appropriate meta-analytical
techniques, the risk of transmission of hepatitis B infection.
METHODS
Search strategy
MEDLINE (1966-March 2011), EMBASE (1980-March 2011), Database of
Abstracts of Reviews of Effects (1991-March 2011), and ACP Journal Club
(1991-March 2011), International Pharmaceutical Index (1970-March 2011),
BIOSIS Previews (1969-March 2011) and Web of Science (1961-March 2011)
were searched to identify the relevant studies and abstracts. The
initial search strategy was developed from the MeSH subject headings
"hepatitis" and "tattoo" in MEDLINE. We reviewed the
titles for their relevance to this study, and then examined subject
headings and abstracts. We searched the proceedings and conference
abstracts through the databases PapersFirst (1993) and ProceedingsFirst
(1993) up to March 2011. Authors' names and year of publication
from key papers were entered into the cited reference search in the Web
of Science. References of the retrieved studies and review articles were
screened for any potentially missed articles. We also hand searched the
reference lists of retrieved studies as well as journals related to
"hepatitis", "hepatology", "transfusion",
"blood", "infection", "epidemiology",
"gastroenterology", and abstracts and books related to
hepatitis. We did not have any language restrictions. Details of the
search strategy are available upon request from the authors.
Selection criteria
Observational studies that assessed the association between
tattooing and hepatitis B were included if they clearly defined: 1)
hepatitis B as either primary or secondary outcome based on serology
test, and 2) tattoos as either primary or secondary exposure; and
presented relative risks or odds ratios (ORs) and their corresponding
95% confidence intervals (95% CI) or provided sufficient data to compute
these parameters. If a study was published in different phases or data
were duplicated in more than one publication, we only included the most
recent. Two authors (SJ and SB) scanned the titles of abstracts
identified through our search strategy and excluded articles that did
not meet the selection criteria, such as those on basic sciences, review
articles, letters to editor, and commentaries.
Data extraction
We created a spreadsheet and recorded study characteristics,
including authors' names, publication year, country of study, study
design, sample size, study population, mean age and/or range, gender of
participants, type of risk factors or confounders adjusted for, outcome
of interest (hepatitis B), and the adjusted OR and 95% CI. Included
articles were reviewed in full by two independent reviewers (SJ and SB).
All discrepancies were resolved after reviewing the source papers and
further discussion among two reviewers. Studies were included only after
full consensus was achieved. For studies that provided several levels of
exposure, each exposure was categorized and analyzed in the designated
subgroup.
Statistical analysis
We pooled the OR and 95% CI of all studies to calculate the overall
risk. If a study provided more than one OR, we used the OR that was
representative of all participants to calculate the overall pooled OR
and 95% CI of all studies included in the meta-analysis. If the overall
OR (95% CI) for all participants was not provided in a study, we
calculated the pooled OR of that study and then used it for the purpose
of calculation of overall OR (95% CI) of our meta-analysis. We performed
several subgroup analyses to investigate the association between
tattooing and risk of hepatitis B among different populations. We
conducted subgroup analyses based on the study population and study
design (case control, cohort, and cross-sectional). We grouped the
studies on tattoos and hepatitis B into four main groups: 1) community
sample (e.g., blood donors, students, and pregnant women), 2) hospital
samples, 3) prisoners and 4) high-risk populations (street youth,
persons with HIV, people who use drugs, those whose tattoos were done
with reused tattoo needles, and those tattooed in non-professional
tattoo parlours). Because only three studies in our review [W5,W18,W31]
* were reporting samples derived from blood donors, we did not create a
subgroup for this group; however, those studies are included in the
pooled analysis. We calculated pooled OR (95% CI) from one study [W9]
that provided data on both tattoos performed in non-professional
parlours and tattoos done with reused needles, and used it in the
subgroup of high-risk subjects.
For all analyses, we weighted the study-specific adjusted log ORs
by the inverse of their variances. A random effects model was used to
estimate the pooled adjusted OR. Statistical heterogeneity between
studies was evaluated with Higgins I2 statistic. (7) Sensitivity
analysis was carried out to assess the influence of individual studies
and then repeating the analysis by excluding the studies with the
largest weights.
[FIGURE 1 OMITTED]
RESULTS
Figure 1 shows the results of our search strategy and step-by-step
inclusion and exclusion of the retrieved papers. We identified a total
of 516 citations related to risk factors of hepatitis. After excluding
273 duplicates, 243 titles were reviewed, of which 64 were selected for
abstract review; of these, 59 studies were selected for full-text
review. The review process resulted in 31 studies (19 cross-sectional, 9
case-control, 3 cohort), with a total of 665,169 participants from 19
countries, being included in the meta-analysis. All studies identified
were in English. Characteristics of the 31 studies included in the
meta-analysis are presented in Table 1.
We found a statistically significant association (pooled OR=1.48,
95% CI: 1.30-1.68; [I.sup.2]=47%) between tattooing and risk of trans
mission of hepatitis B infection. In subgroup analysis, we found the
strongest association between tattooing and risk of hepatitis B for
samples derived from high-risk groups (OR=1.64, 95% CI: 1.32-2.03;
[I.sup.2]=0%), followed by community samples (OR=1.47, 95% CI:
1.12-1.92; [I.sup.2]=58%), hospital samples (OR=1.45, 95% CI: 1.07-1.97;
[I.sup.2]=30%), and prison samples (OR=1.30, 95% CI: 1.01-1.66;
[I.sup.2]=56%). Figure 2 shows the results of our meta-analysis for
association between tattooing and risk of hepatitis B for four main
subgroups of samples derived from community, hospital, prison inmates,
and high-risk groups.
We conducted a subgroup analysis to investigate the effect of the
study design on the association of tattooing and the risk of hepatitis
B. The association between tattooing and risk of hepatitis B was the
strongest among case-control studies (OR=1.97, 95% CI: 1.45-2.69;
[I.sup.2]=46%), followed by cohort (OR=2.01, 95% CI: 1.522.67;
[I.sup.2]=0%) and cross-sectional (OR=1.34, 95% CI: 1.16-1.54;
[I.sup.2]=43%) studies. We also conducted sensitivity analysis to review
the effect of three studies with wide confidence intervals [W2,W12,W31]
on overall pooled OR (95% CI). The analysis was conducted multiple
times; first, all three studies were removed from the analysis, and then
one study at a time was removed. We did not find a significant
difference in OR (95% CI) when all three studies were removed from the
analysis (OR=1.44, 95% CI: 1.20-2.08; [I.sup.2]=38%). We also found no
significant difference between the pooled pre-sensitivity effect size
(OR=1.48, 95% CI: 1.30-1.68; [I.sup.2]=47%) and post-sensitivity effect
size after removing any of these studies (OR=1.46, 95% CI: 1.29-1.65;
[I.sup.2]=41%) [W2], (OR=1.45, 95% CI: 1.29-1.64; [I.sup.2]=41%) [W12],
(OR=1.48, 95% CI: 1.30-1.68; [I.sup.2]=46%) [W31] from the analysis.
In studies of tattooing and hepatitis B, a moderate heterogeneity
was detected ([I.sup.2]=47%) in this review. Using Jackknife method, (8)
six studies [W1,W2,W3,W10,W12,W13] were identified to be the source of
heterogeneity. Removing these studies from the analysis resulted in
elimination of the heterogeneity from pooled and subgroup analysis.
However, pre- and post-sensitivity overall pooled OR (95% CI)
(pre-sensitivity OR=1.48, 95% CI: 1.30-1.68, [I.sup.2]=47%;
post-sensitivity OR=1.58, 95% CI: 1.44-1.74, [I.sup.2]=0%), and
pre-/post-sensitivity OR (95% CI) in community samples (pre-sensitivity
OR=1.47, 95% CI: 1.12-1.92, [I.sup.2]=58%; post-sensitivity OR=1.51, 95%
CI: 1.25-1.84, [I.sup.2]=0%) and prisoners (pre-sensitivity OR=1.30, 95%
CI: 1.01-1.66, [I.sup.2]=56%; post-sensitivity OR=1.62, 95% CI:
1.33-1.96, [I.sup.2]=0%) increased, whereas for hospital samples
(pre-sensitivity OR=1.45, 95% CI: 1.07-1.97, [I.sup.2]=16%;
post-sensitivity OR=1.35, 95% CI: 1.07-1.72, [I.sup.2]=0%), these
diminished. Differences in the study populations, time and place that
the studies were conducted and the study methodology would potentially
contribute to the heterogeneity in this study. To further examine the
effect of adjustment done in some studies, we removed studies with
non-adjusted OR from all subgroups. We did not find any significant
change in pooled OR (95% CI) (pre-sensitivity OR=1.48, 95% CI:
1.30-1.68, [I.sup.2]=47%; post-sensitivity OR=1.48, 95% CI: 1.24-1.77,
[I.sup.2]=30%). We did not conduct such a sensitivity for subgroups
because exclusion of studies with unadjusted OR resulted in a small
number of studies in each subgroup.
DISCUSSION
Results of our systematic review indicate an increased risk of
hepatitis B infection among those who receive tattoos. The risk is
persistent among all population subgroups. Not surprisingly, the
strongest association is observed among samples from the high-risk
populations: injection drug users, sex workers, street youth, HIV
patients, those whose tattoos were done with reused tattoo needles, and
those tattooed in non-professional tattoo parlours.
A major strength of our review is the large number of studies and
the multinational nature of the study sample. We found an association
between tattooing and hepatitis B in all subgroups and across all study
designs. Our findings are consistent with the literature which documents
an association between tattooing and risk of transmission of other
infections (including HIV, (9) leprosy, (10) tetanus, (9) and
Methicillin Resistant Staphylococcus Aureus (11)) and with a systematic
review of hepatitis C. (12) Several studies have indicated a
dose-response relationship between tattooing and the risk of
transmission of hepatitis C. (13-15) In fact, the risk of transmission
of hepatitis C infection increases with the increase in the surface area
covered by a tattoo as well as the number of tattoos received by an
individual. Taking into account that the infectivity and inocula differ
between HBV and HCV and considering that such dose-response associations
between tattooing and risk of transmission of hepatitis B were not
reported in the included studies, this topic requires further study.
The risk from tattooing may depend on the background prevalence of
hepatitis B infection and immunization uptake in the subgroups of a
population. Tattooing among prisoners is an area of concern due to the
high prevalence of hepatitis B infection among incarcerated individuals;
the background rates of hepatitis B infection are 10 to 20 times higher
among prisoners (16-18) than among the general public. Reusing and
sharing tattoo needles are reported to be common practice among almost
45% of prisoners. (19) However, our results indicate that the
association between tattooing and the risk of transmission of hepatitis
B among prisoners is minimally lower compared with other subgroups. This
may be because of variations in study design and sampling method among
included studies or immunization programs within correctional
facilities. The results of this study support the recommendations of
experts, (19) advocating the need for prison programs that provide safer
tattooing practices to inmates. Such programs may also prevent the
transmission of other blood-borne infections among the prison
population.
Given that hepatitis can spread through percutaneous or mucous
membrane exposure to blood, (20) needlestick injury, (21) and tattooing
using unsterile equipment, (22) several measures can be implemented to
prevent the transmission of hepatitis B among tattoo recipients.
Educational programs for tattoo parlour owners and tattoo artists which
reinforce the guidelines and emphasize the importance of appropriate
infection control measures should be implemented. These measures include
the use of single-use sterile tattoo needles, proper functioning of
autoclaves, use of appropriate disinfectants and keeping records of
sterilization techniques. Regular and unscheduled inspection of tattoo
parlours conducted by health protection units of the local health
authorities may improve adherence to the current guidelines.
Insufficient follow-up can cause public panic and has resulted in
lawsuits against health authorities and business owners in Canada and
the United States in recent years. (23)
Tattoo parlours should be required to keep records of their clients
and to report any adverse event related to tattooing to local health
authorities. (12,22) Tattoo recipients should be provided with written
handouts informing them about the risks related to tattooing and the
signs and symptoms of common infections that can be transmitted through
tattooing. Last, tattoo artists and clients should be aware of the
availability of an effective vaccine for the prevention of hepatitis B
infection.
Many countries have introduced a universal infant and/or adolescent
hepatitis B immunization program; therefore those who undergo tattooing
at a young age are now likely to be immunized. In Canada, all
unimmunized individuals who are at increased risk of hepatitis B
infection (i.e., IDU, persons with high-risk sexual behaviour) are
recommended to receive hepatitis B vaccine. However, this strategy has
several limitations. Risk factors cannot be identified for about 25% of
acute hepatitis B infections, (24) and there is poor compliance with the
hepatitis B vaccine schedule in risk groups. (25) Therefore, it is
important to ensure that individuals have been fully
immunized--including those who move between jurisdictions which may have
different immunization schedules--and to concentrate on immunization of
high-risk and prison populations.
Over time, there has been a shift in the demographics of the people
who acquire tattoos. Tattoos are no longer popular just among people
with high-risk behaviour, but are becoming common among younger adults
with a more conservative lifestyle. This increased prevalence of
tattooing warrants the need for prevention programs that target young
adults. (26)
Our study is subject to several limitations due to the
observational nature of the studies included in the review. Information
on the past history of tattooing may not reflect the current population
risk of hepatitis infection. Second, recall bias and social desirability
bias may affect the validity of the information collected in
observational studies. Finally, it is important to note that background
prevalence of hepatitis B infection and hepatitis B vaccinations vary in
different settings and countries. We did not have specific geographic
data on the transmission of hepatitis B. Further studies to determine
the pre-tattoo hepatitis B serostatus and the prevalence of other
high-risk behaviours are needed to fully assess the association between
tattooing and hepatitis B transmission.
The results of this systematic review are consistent with an
increase in all population groups in the risk of contracting hepatitis B
through tattooing. Immunization of susceptible individuals with
hepatitis B vaccine, unscheduled inspection and enforcement of infection
control guidelines in tattoo parlours, and provision of safe tattoo
programs to inmates can play an important role in the interruption of
transmission.
Conflict of Interest: None to declare.
Received: September 20, 2011
Accepted: January 21, 2012
REFERENCES
(1.) Laumann AE, Derick AJ. Tattoos and body piercing in the United
States: A national data set. J Am Acad Dermatol 2006;55:413-21.
(2.) Deschesnes M, Demers S, Fines P. Prevalence and
characteristics of body piercing and tattooing among high school
students. Can J Public Health 2006;97(4):325-29.
(3.) Health Canada. Special Report on Youth, Piercing, Tattooing
and Hepatitis C TrendScan Findings, Health Canada, March 2001. Available
at: http://www.phac-aspc.gc.ca/hepc/pubs/
youthpt-jeunessept/pdf/youth_piercings.pdf (Accessed August 12, 2009).
(4.) Glasgow KW, Schabas R, Williams DC, Wallace E, Nalezyty LA. A
population-based hepatitis B seroprevalence and risk factor study in a
Northern Ontario town. Can J Public Health 1997;88(2):87-90.
(5.) Mast E, Mahoney F, Kane M, Margolis HS. Hepatitis B vaccine.
In: Plotkin SA, Orenstein WA (Eds.), Vaccines, 4th ed. Philadelphia, PA:
WB Saunders, 2004.
(6.) Brooks EA, Lacey LF, Payne SL, Miller DW. Economic evaluation
of lamivudine compared with interferon-alpha in the treatment of chronic
hepatitis B in the United States. Am J Manag Care 2001;7(7):677-82.
(7.) Higgins JP, Thompson SG. Quantifying heterogeneity in a
meta-analysis. Stat Med 2002;21:1539-58.
(8.) Efron B. Nonparametric estimates of standard error: The
jackknife, the bootstrap and other methods. Biometrika 1981;68:589-99.
(9.) Nishioka S de A, Gyorkos TW. Tattoos as risk factors for
transfusion-transmitted diseases. Int J Infect Dis 2001;5(1):27-34.
(10.) Ghorpade A. Inoculation (tattoo) leprosy: A report of 31
cases. J Eur Acad Dermatol Venereol 2002;16:494-99.
(11.) United States Centers for Disease Control and Prevention
(CDC). Methicillin-resistant Staphylococcus aureus skin infections among
tattoo recipients--Ohio, Kentucky, and Vermont, 2004-2005. MMWR
2006;55(24):677-69.
(12.) Jafari S, Copes R, Baharlou S, Etminan M, Buxton J. Tattooing
and risk of transmission of hepatitis C: A systematic review and
meta-analysis. Int J Infect Dis 2010;14(11):e928-40.
(13.) Roy E, Haley N, Leclerc P, Boivin JF, Cedras L, Vincelette J.
Risk factors for hepatitis C virus infection among street youths. CMAJ
2001;165:557-60.
(14.) Haley RW, Fischer RP. Commercial tattooing as a potentially
important source of hepatitis C infection: Clinical epidemiology of 626
consecutive patients unaware of their hepatitis C serologic status.
Medicine 2001;80:134-51.
(15.) Kim YS, Ahn YO, Lee HS. Risk factors for hepatitis C virus
infection among Koreans according to the hepatitis C virus genotype. J
Korean Med Sci 2002;17:187-92.
(16.) Weild AR, Gill ON, Bennett D, Livingstone SJM, Parry JV,
Curry L. Prevalence of HIV, hepatitis B, and hepatitis C antibodies in
prisoners in England and Wales: A national survey. Commun Dis Public
Health 2000;3:121-26.
(17.) Khan AJ, Simard EP, Bower WA, Wurtzel HL, Khristova M, Wagner
KD, et al. Ongoing transmission of hepatitis B virus infection among
inmates at a state correctional facility. Am J Public Health
2005;95:1793-99.
(18.) Canada Dept of Health and Welfare. Seroepidemiologic study of
hepatitis B and C viruses in federal correctional institutions in
British Columbia. Can Dis Wkly Rep 1990;52:265-66.
(19.) Crofts N, Stewart T, Hearne P, Ping XY, Breschkin AM,
Locarnini SA. Spread of blood borne viruses among Australian prison
entrants. BMJ1995;310:285 88.
(20.) Gerberding JL. Incidence and prevalence of human
immunodeficiency virus, hepatitis B virus, hepatitis C virus, and
cytomegalovirus among health care personnel at risk for blood exposure:
Final report from a longitudinal study. J Infect Dis 1994;170:1410-17.
(21.) Sartori M, La Terra G, Aglietta M, Manzin A, Navino C,
Verzetti G. Transmission of hepatitis C via blood splash into
conjunctiva. (Letter) Scand J Infect Dis 1993;25:270-7l.
(22.) Kiyosawa K, Sodeyama T, Tanaka E, Nakano Y, Furuta S,
Nishioka K, et al. Hepatitis C in hospital employees with needle stick
injuries. Ann Intern Med 1991;115:367-69.
(23.) City TV. Tattoo Parlour, Peel Sued Over Possible HIV,
Hepatitis Exposure. March 19, 2009. Available at:
http://www.citytv.com/toronto/citynews/news/
local/article/9477--tattoo-parlour-peel-sued-
over-possible-hiv-hepatitis-exposure (Accessed August 19, 2011).
(24.) Boulos D, Goedhuis NJ, Wu J, Baptiste B, Poliquin D, Furseth
J, et al. Enhanced surveillance for acute and likely acute hepatitis B
in Canada: 1999 to 2002. Can J Infect Dis Med Microbiol 2005;16:275-81.
(25.) Sellors J, Zimic-Vincetic M, Howard M, Chernesky MA. Lack of
compliance with hepatitis B vaccination among Canadian STD clinic
patients: Candidates for an accelerated immunization schedule? Can J
Public Health 1997;88:210 11.
(26.) Public Health Agency of Canada. Special Report on Youth,
Piercing, Tattooing and Hepatitis C Trendscan Findings. Available at:
http://www.phacaspc.gc.ca/hepc/pubs/youthpt-
jeunessept/pdf/youth_piercings.pdf (Accessed August 26, 2011).
Siavash Jafari, MD, MHSc, [1] Jane A. Buxton, MBBS, FRCPC, [1]
Kourosh Afshar, MD, MHSc, FRCPC FAAP, [2] Ray Copes, MD, MSc, [2] Souzan
Baharlou, MD [3]
Author Affiliations
[1.] School of Population and Public Health, Faculty of Medicine,
University of British Columbia, Vancouver, BC
[2.] Division of Pediatric Urology, BC Children's Hospital,
Vancouver, BC
[3.] Director, Environmental and Occupational Health, Ontario
Agency for Health Protection and Promotion, Toronto, ON
Correspondence: Dr. Jane A. Buxton, Epidemiologist and Harm
Reduction Lead, BC Centre for Disease Control, 655 West 12th Avenue,
Vancouver, BC V5Z 4R4, Tel: 604-707-2573, Fax: 604-707-2516, E-mail:
jane.buxton@ubc.ca
* See Table 1 for a list of the 31 studies (W1 through W31).
Table 1. Characteristics of Studies Included in Both Systematic Review
and Meta-analysis
Study Author Year Location Sample Size
ID (case/control)
W1 Anda 1985 USA 619
W2 Auamnoy 2003 Thailand 92 (46/46)
W3 Babudieri 2005 Italy 973
W4 Butler 2007 Australia 612
W5 Christensen 2000 Denmark 325
W6 Christensen 2001 Denmark 10,862
W7 Coppola 2007 Italy 3579
W8 Hull 1985 USA 455
W9 Hwang 2006 USA 7960
Hwang 2006 USA 7960
Hwang 2006 USA 7960
Hwang 2006 USA 7960
Hwang 2006 USA 7960
Hwang 2006 USA 7960
Hwang 2006 USA 7960
W10 Hymas 1989 Sudan 851
W11 Khan 2005 USA 1124
W12 Ko 1990 Taiwan 90/180
W13 Lai 2007 Taiwan 286
W14 Liao 2006 Taiwan 298
W15 Luksamijarulkul 2008 Thailand 354
W16 Mariano 2004 Italy 2964/7221
W17 Mele 1995 Italy 5241 (363/4879)
W18 Nishioka 2003 Brazil 345
W19 Nurgalieva 2007 Kazakhstan 290
W20 Pallas 1999 Spain 1215
W21 Pereira 2006 Brazil 1025
W22 Phoon 1988 Singapore 6328
W23 Pourahmad 2007 Iran 1431 (497/934)
W24 Roy 1999 Canada 437
W25 Sali 2005 Iran (500/434)
W26 Samuel 2001 USA 2898
Samuel 2001 USA 2898
W27 Sebastian 1992 Brunei (187/187)
W28 Shi 2007 Taiwan (476/1421)
W29 Tawk 2005 Australia 2120
W30 Vahid 2005 Iran 454/428
W31 Wada 1999 Japan 95
Wada 1999 Japan 95
Study Study Sample Age, Range,
ID Design Derived From Mean (SD)
W1 CS * Prison Mean (24.8-28.2)
W2 CC ([dagger]) Hospital 22.88 [+ or -] 1.41
W3 CS Prison 30-44 (36.9)
W4 CS Prison All ages
W5 CO ([double Prison All ages
dagger])
W6 CS Blood donors Median=48
W7 CS Healthy subjects 33.19 (12.18)
W8 CS Prison 18-71
W9 CS Students; all >18; median age:
21.62
CS 1-2 tattoo(s) Not reported
CS >3 tattoos Not reported
CS Professional Not reported
tattoo parlour
CS Non-professional Not reported
tattoo parlour
CS New or autoclaved Not reported
needle
CS Reused needle Not reported
W10 CS Community 1-89 (24.6)
W11 CS Prison 18-72
W12 CC Healthy adults Not reported
W13 CS Prison Not reported
W14 CS Prison Not reported
W15 CS Public cleaners 39.5 (7.7)
W16 CC IVDU & blood 15-55
transfusion
W17 CC Community Not reported
W18 CS Blood donors 18-62
W19 CS Community 10-65
W20 CS Prison 30.6 (9.9)
W21 CS HIV patients 17-77
W22 CS Hospital 35-65
W23 CC Prison 25-60
W24 CO Street youth 19.5
W25 CC Hospital 37.6 (15.1)
W26 CO IDU, tattooed >15
in prison
CO IDU, tattooed in >17
community
W27 CC Hospital 20-70
W28 CC Military All 20 years old
W29 CS Hospital 52.3
W30 CC Blood donors HBsAg+ 36.69 (0.82);
HBsAg- 30.88 (0.80)
W31 CS Hospital 24.4 (6.1)
CS Hospital 24.4 (6.1)
Study Gender OR/ 95% CI Adjustment
ID RR
W1 M 0.53 0.31-0.92 IVDU; IVDU and previous
imprisonment; prior hepatitis
or jaundice; race; age
W2 M/F 15.9 1.97-128.16 No
W3 M/F 1.14 0.89-1.47 Age; gender; area of origin;
exposure category; unprotected
sex; transfusions;
imprisonment
W4 M/F 1.66 1.01-2.74 No
W5 M 1.00 0.30-2.90 Age, times in prison, duration
of IDU, sexual risk index, IDU
in prison
W6 M/F 2.89 1.51-5.50 No
W7 M/F 1.85 1.18-2.90 No
W8 M 1.30 0.40-2.60 No
W9 M/F 0.96 0.73-1.25 No
M/F 0.99 0.74-1.32 No
M/F 0.87 0.50-1.50 No
M/F 0.93 0.70-1.23 No
M/F 1.24 0.69-2.23 No
M/F 0.87 0.65-1.18 No
M/F 1.91 1.11-3.30 No
W10 M/F 1.58 1.04-2.39 No
W11 M 1.46 1.04-2.06 Age, race, history of IDU/
STI, lifetime number of female
partners, incarcerated more
than 14 years, tattoo during
incarceration
W12 M 8.10 1.90-34.8 No
W13 0.66 0.32-1.39 No
W14 1.49 0.78-2.84 No
W15 M/F 1.17 0.69-1.99 No
W16 M/F 1.70 1.00-3.10 Sex, age, education,
geographical area, surgical
intervention, dental therapy,
number of sexual partners,
households or sexual partners
of HBsAg/HCV chronic carrier
W17 M/F 2.12 1.10-4.09 Age, sex, education level,
geographic area, surgical
interventions, dental therapy,
sexual exposure
W18 M/F 1.90 0.94-3.83 Syphilis, Chagas' disease, HIV
infection, at least one marker
for HBV, HCV, or HIV
W19 M/F 1.26 0.51-3.12 No
W20 M/F 1.70 0.80-3.5 Age, sex, educational level,
number of sexual partners
W21 M/F 1.60 1.10-2.40 Age, gender, number of sexual
partners
W22 M 1.16 0.75-1.79 Age
W23 M 1.85 1.00-3.43 No
W24 M/F 1.60 0.60-4.20 Age, IDU, number of sexual
partners, body piercing
W25 M/F 1.40 0.85-2.29 No
W26 M/F 2.30 1.40-3.8 Age, study site, race, share
injection equipment, use of
heroin, years of injection
M/F 1.60 1.10-2.50 No
W27 M 2.01 1.05-3.95 Matched case-control
W28 M 1.37 0.98-1.93 No
W29 M/F 1.33 0.88-2.02 No
W30 M/F 4.60 1.50-13.9 No
W31 M/F 1.60 0.00-24.00 No
M/F 1.40 0.37-4.99 No
* CS = Cross-sectional; ([dagger]) CC = Case-control; [double dagger])
CO = Cohort; M = Male; F = Female
Figure 2. Forest plot of meta-analysis of tattooing and risk of
transmission of hepatitis B
log
Study or Subgroup [Odds Ratio] SE Weight
1.4.1 Tattoo and Hepatitis-B, Community sample
Coppola, 2007 0.615 0.23 4.2%
Hwang, 2006 -0.04 0.136 6.4%
Hymas, 1989 0.455 0.213 4.5%
Ko, 1990 2.092 0.741 0.7%
Luksamijarulkul, 2008 0.157 0.27 3.5%
Mele, 1995 0.7514 0.335 2.6%
Nurgalieva, 2007 0.2311 0.462 1.6%
Shi, 2007 0.3148 0.174 5.4%
Subtotal (95% CI) 29.0%
Heterogeneity: [Tau.sup.2] = 0.08; [Chi.sup.2] = 16.55, df = 7
(P = 0.02); [I.sup.2] = 58%
Test for overall effect: Z = 2.77 (P = 0.006)
1.4.3 Tattoo and Hepatitis-B, Hospital samples
Auamnoy, 2003 2.7663 1.0648 0.4%
Phoon, 1988 0.1484 0.22 4.4%
Sali, 2005 0.3365 0.25 3.8%
Sebastian, 1992 0.6981 0.345 2.5%
Tawk, 2005 0.2582 0.228 4.2%
Wada, 1999 0.355 0.59 1.1%
Subtotal (95% CI) 16.4%
Heterogeneity: [Tau.sup.2] = 0.04; [Chi.sup.2] = 7.16, df = 5
(P = 0.21); [I.sup.2] = 30%
Test for overall effect: Z = 2.38 (P = 0.02)
1.4.4 Tattoo and Hepatitis-B, Prisoners
Anda, 1985 -0.6349 0.28 3.3%
Babudieri, 2005 0.13 0.13 6.6%
Butler, 2004-b 0.505 0.256 3.7%
Christensen, 2001 0.001 0.543 1.2%
Hull, 1985 0.2624 0.353 2.4%
Khan, 2005 0.3784 0.175 5.4%
Lai, 2007 -0.4155 0.379 2.2%
Liao, 2006(2) 0.3988 0.329 2.7%
Pallas, 1999 0.528 0.37 2.3%
Pourahmad, 2007 0.6152 0.315 2.9%
Samuel, 2001 0.8329 0.256 3.7%
Subtotal (95% CI) 36.4%
Heterogeneity: [Tau.sup.2] = 0.09; [Chi.sup.2] = 22.96, df = 10
(P = 0.01); [I.sup.2] = 56%
Test for overall effect: Z = 2.07 (P = 0.04)
1.4.5 Tattoo and Hepatitis-B, Highr risk group and IDUs
Hwang, 2006 0.523 0.195 4.9%
Mariano, 2004 0.5306 0.307 3.0%
Pereira, 2006 0.47 0.206 4.7%
Roy, 1999 0.47 0.492 1.4%
Samuel, 2001 0.473 0.227 4.2%
Subtotal (95% CI) 18.3%
Heterogeneity: [Tau.sup.2] = 0.00; [Chi.sup.2] = 0.06, df = 4
(P = 1.00); [I.sup.2] = 0%
Test for overall effect: Z = 4.54 (P < 0.00001)
Total (95% CI) 100.0%
Heterogeneity: [Tau.sup.2] = 0.05; [Chi.sup.2] = 50.42, df = 29
(P = 0.008); [I.sup.2] = 42%
Test for overall effect: Z = 5.45 (P < 0.00001)
Test for subgroup differences: [Chi.sup.2] = 2.04, df = 3 (P = 0.56),
[I.sup.2] = 0%
Odds Ratio
Study or Subgroup IV, Random, 95% CI
1.4.1 Tattoo and Hepatitis-B, Community sample
Coppola, 2007 1.85 [1.18, 2.90]
Hwang, 2006 0.96 [0.74, 1.25]
Hymas, 1989 1.58 [1.04, 2.39]
Ko, 1990 8.10 [1.90, 34.62]
Luksamijarulkul, 2008 1.17 [0.69, 1.99]
Mele, 1995 2.12 [1.10, 4.09]
Nurgalieva, 2007 1.26 [0.51, 3.12]
Shi, 2007 1.37 [0.97, 1.93]
Subtotal (95% CI) 1.47 [1.12, 1.92]
Heterogeneity: [Tau.sup.2] = 0.08; [Chi.sup.2] = 16.55, df = 7
(P = 0.02); [I.sup.2] = 58%
Test for overall effect: Z = 2.77 (P = 0.006)
1.4.3 Tattoo and Hepatitis-B, Hospital samples
Auamnoy, 2003 15.90 [1.97, 128.16]
Phoon, 1988 1.16 [0.75, 1.79]
Sali, 2005 1.40 [0.86, 2.29]
Sebastian, 1992 2.01 [1.02, 3.95]
Tawk, 2005 1.29 [0.83, 2.02]
Wada, 1999 1.43 [0.45, 4.53]
Subtotal (95% CI) 1.45 n.07, 1.971
Heterogeneity: [Tau.sup.2] = 0.04; [Chi.sup.2] = 7.16, df = 5
(P = 0.21); [I.sup.2] = 30%
Test for overall effect: Z = 2.38 (P = 0.02)
1.4.4 Tattoo and Hepatitis-B, Prisoners
Anda, 1985 0.53 [0.31, 0.92]
Babudieri, 2005 1.14 [0.88, 1.47]
Butler, 2004-b 1.66 [1.00, 2.74]
Christensen, 2001 1.00 [0.35, 2.90]
Hull, 1985 1.30 [0.65, 2.60]
Khan, 2005 1.46 [1.04, 2.06]
Lai, 2007 0.66 [0.31, 1.39]
Liao, 2006(2) 1.49 [0.78, 2.84]
Pallas, 1999 1.70 [0.82, 3.50]
Pourahmad, 2007 1.85 [1.00, 3.43]
Samuel, 2001 2.30 [1.39, 3.80]
Subtotal (95% CI) 1.30 [1.01, 1.66]
Heterogeneity: [Tau.sup.2] = 0.09; [Chi.sup.2] = 22.96, df = 10
(P = 0.01); [I.sup.2] = 56%
Test for overall effect: Z = 2.07 (P = 0.04)
1.4.5 Tattoo and Hepatitis-B, Highr risk group and IDUs
Hwang, 2006 1.69 [1.15, 2.47]
Mariano, 2004 1.70 [0.93, 3.10]
Pereira, 2006 1.60 [1.07, 2.40]
Roy, 1999 1.60(0.61,4.20]
Samuel, 2001 1.60 [1.03, 2.50]
Subtotal (95% CI) 1.64 [1.32, 2.03]
Heterogeneity: [Tau.sup.2] = 0.00; [Chi.sup.2] = 0.06, df = 4
(P = 1.00); [I.sup.2] = 0%
Test for overall effect: Z = 4.54 (P < 0.00001)
Total (95% CI) 1.42 [1.25, 1.61]
Heterogeneity: [Tau.sup.2] = 0.05; [Chi.sup.2] = 50.42, df = 29
(P = 0.008); [I.sup.2] = 42%
Test for overall effect: Z = 5.45 (P < 0.00001)
Test for subgroup differences: [Chi.sup.2] = 2.04, df = 3 (P = 0.56),
[I.sup.2] = 0%
