Abstract
Background
Tattooing can deliver carcinogens directly into the skin and cause immunological responses, and yet the relationship between tattooing and melanoma risk is unknown.
Methods
In a population-based case-control study with 1167 melanoma cases (566 in situ; 601 invasive) and 5835 frequency-matched controls, we examined tattooing and melanoma risk using multivariable logistic regression to calculate odds ratios (ORs) and 95% confidence intervals (CIs).
Results
Although ever receiving a tattoo was not strongly associated with melanoma risk, heavier tattooing exposure was associated with decreased risk. Overall melanoma risk was decreased among individuals who had received 4 or more tattoo sessions (OR = 0.44, 95% CI = 0.27 to 0.67) and individuals who had 3 or more large tattoos (OR = 0.26, 95% CI = 0.10 to 0.54) compared with those who were never tattooed. Invasive melanoma risk was also decreased among individuals who received their first tattoo before age 20 (OR = 0.48, 95% CI = 0.29 to 0.82) compared with never tattooed individuals.
Conclusions
Our findings suggest that more tattoo exposure is associated with reduced melanoma risk, which does not support previously hypothesized associations between tattooing and increased melanoma risk. Unmeasured confounding is likely to contribute to our findings because we were not able to control for important melanoma risk factors. Potential causes of these associations could include sun exposure-related behaviors or immune responses to tattooing. Further investigation is warranted to clarify these relationships.
Introduction
Almost one third of adults in the United States have a tattoo,1 but the long-term health effects of tattooing are largely unknown. Many commercially available tattoo inks contain carcinogens including metals, polycyclic aromatic hydrocarbons, and primary aromatic amines.2 In addition, inks can photodegrade when exposed to ultraviolet radiation and form new toxic compounds.3-5 Rather than remaining fully in the skin, inks and their carcinogenic components accumulate in regional lymph nodes.6,7 Tattooing can also result in long-term inflammatory and immune responses because allergies from tattooing can occur months to years after tattooing.8,9 These immune effects may be relevant to melanoma risk because most melanoma tumors are highly mutagenic and elicit complex immune responses.10 In case reports, more than 160 skin cancer cases have been observed within tattoos, including at least 43 cases of melanoma.11 However, only 1 epidemiological study of tattooing and melanoma risk has been published to date that observed increased risk of combined melanoma and nonmelanoma skin cancer associated with ever getting tattooed and ever receiving a large tattoo.12 To our knowledge, no published studies to date have examined whether tattooing may increase melanoma risk alone.
Patterns of tattooing prevalence by sex have shifted over time in the United States. In 2003, the proportion of individuals with tattoos was similar between women (15%) and men (16%).13 However, by the early 2010s, this pattern began to diverge with 23% of women and 19% of men having a tattoo in 2012.13 As of 2023, 38% of women and 27% of men in the United States were estimated to have a tattoo.1 Among both women and men, individuals with tattoos are more likely to be younger, to be not married, to have a lower level of education, and to smoke than those without tattoos.1,14,15 Tattooing prevalence also varies by race and ethnicity, with a higher prevalence of tattoos among Hispanic, American Indian/Alaskan Native, and Pacific Islander individuals compared with non-Hispanic White individuals.14
Melanoma risk increases with age, but patterns of melanoma incidence by age vary between women and men.16 In the US population, women have a higher risk of melanoma than men before the age of 45 years,16 which may partly reflect a greater tendency toward overdiagnosis in women.17 Conversely, after age 45, the risk of melanoma is higher among men than among women.16 Melanoma mortality also increases with age. Between 1999 and 2020 in the United States, individuals ages 65 and older had a 12.2 per 100 000 age-adjusted mortality rate, whereas the rate among ages 45-64 was 3.3 and among ages 25-44 was 0.8 per 100 000, respectively.18 Between 2017 and 2021, the age-adjusted incidence rate of melanoma in Utah was approximately 84% higher and mortality was approximately 29% higher than the rest of the United States.19 During this period, the age-standardized incidence of melanoma in Utah was higher among men (50.4 per 100 000) than women (34.3 per 100 000), as was melanoma mortality (3.8 per 100 000 among men and 1.9 per 100 000 among women).19
In this study, we evaluated associations between tattoo exposures and melanoma incidence in Utah, the state with the highest melanoma incidence in the United States.20 Because patterns of tattooing and melanoma incidence differ considerably between women and men, in addition to overall patterns we examined associations separately by sex.
Methods
We conducted a population-based case-control study. A flowchart of the inclusion/exclusion process is shown in Figure 1. All incident cases of in situ and invasive melanoma diagnosed in Utah between January 1, 2020, and June 30, 2021, ages 19-79 years old were identified by the Utah Cancer Registry using rapid case ascertainment, and a letter was mailed inviting them to participate in the study. The University of Utah Institutional Review Board (IRB) determined this research exempt (IRB #00123466) and waived the need for documentation of informed consent. A trained interviewer attempted to contact each case via telephone using a standard protocol and informed individuals that by completing the interview they were consenting to participate in the study. Of the 3032 incident cases, 133 were determined to be outside of the interviewing time frame ending in January 2024 for a total of 2899 cases included in recruitment. Of these, 105 (4%) were deceased before first contact, 1111 (38%) were unable to be reached, and 505 (17%) refused. Telephone surveys were completed by 1178 individuals for a response proportion of 41%. Individuals were excluded after interview if they were not a Utah resident 1 year before diagnosis (n = 5) or were missing tattoo data (n = 6). The final analytic dataset included 1167 melanoma cases (566 in situ and 601 invasive). Interviewers captured participants’ responses in REDCap housed at the University of Utah Clinical and Translational Science Institute. Cases were asked to think back to the time period 1 year before diagnosis when answering the survey questions. The full questionnaire administered to cases is available in the Supplementary Material. We used the piping feature in REDCap to fill in questions with the time period relevant to each participant’s diagnosis.
Figure 1.
Flowchart of included melanoma cases and frequency-matched controls in Utah. Abbreviation: BRFSS = Behavioral Risk Factor Surveillance System.
Controls were selected from respondents to the 2020-2021 Utah Behavioral Risk Factor Surveillance System (BRFSS) survey. As previously described,21 we worked with the Utah Department of Health and Human Services to add 3 tattoo questions to the BRFSS survey: (1) What is the total number of tattooing sessions you have had? (2) How many of your tattoos are bigger than your palm? and (3) How old were you when you got your first tattoo? Participants were asked to include every tattoo they had ever received that was administered using a tattoo machine. Surveys were completed by 21 542 participants with response proportions of 55% in 2020 and 47% in 2021. We excluded individuals with unknown age (n = 367), individuals above age 78 years (n = 1805) to be comparable to cases who were ages 19-79, as cases were asked about the time period 1 year before diagnosis, individuals who reported a prior melanoma diagnosis (n = 474), and individuals who were missing tattoo data (n = 2508), leaving 16 388 participants eligible for control selection. We frequency-matched controls to cases in a 5:1 ratio on sex, 5-year age group, and race and ethnicity (Hispanic, non-Hispanic American Indian or Alaska Native, non-Hispanic Asian, non-Hispanic Black, non-Hispanic Pacific Islander, non-Hispanic White, non-Hispanic multiracial, non-Hispanic other, unknown). In total, 5835 controls were selected.
Statistical analysis
We computed odds ratios (ORs) and 95% confidence intervals (CIs) from logistic regression models associating tattoo exposures (ever tattooed, time since first tattoo, age at first tattoo, number of tattoo sessions, number of large tattoos) with any melanoma diagnosis, and separately for in situ and invasive melanoma. We created a directed acyclic graph (DAG) using the DAGitty browser-based tool (dagitty.net) to identify potential confounders for model adjustment (Figure S1). Models were adjusted for sex, age (5-year groups), race and ethnicity (non-Hispanic White, Hispanic, all other racial and ethnic groups), education (<high school diploma, high school diploma, some college, 4-year college degree or more), ever smoking (yes/no), physical activity in the past 30 days (yes/no), and body mass index (BMI; <25/25+ kg/m2). We then fit models stratified by sex. Because both tattooing prevalence and melanoma risk vary by race and ethnicity, we also conducted a sensitivity analysis restricted to non-Hispanic White individuals.
Data on the melanoma risk factors indoor tanning, sunscreen use on a sunny summer day, having had at least 1 red or painful sunburn in the 1- to 2-year period before diagnosis, tattoo sun exposure in the summer, ability to tan, hair and eye color, and personal and family history of melanoma were collected for cases but were not available for controls because of limits on the number of questions that could be added to the BRFSS. In supplemental analyses, we computed frequencies of these factors in cases, among those who never received a tattoo, those with 1 tattoo session, and those with 2 or more tattoo sessions.
We used propensity score models to explore potential confounders in the relationships between ever tattooing and number of tattoo sessions with melanoma risk.22 We first constructed models separately to predict ever receiving a tattoo, 1 tattoo session, and 2 or more tattoo sessions compared with never receiving a tattoo using demographic, health, and risk-taking variables within the BRFSS control population. We took an iterative approach by dropping variables, collapsing categories, and considering 2-way interactions while using the Akaike information criterion to select the optimal model. We then used comparable models including all variables available among cases and controls to match exposed and nonexposed individuals in a 10:1 ratio. We used nearest neighbor matching with a caliper of 0.2 to improve covariate balance using the MatchIt package in R.23 We fit logistic regression models to calculate odds ratios and 95% confidence intervals within the propensity score-matched sets associating tattoo exposures with melanoma risk.
Because this study took place in Utah where more than half of the population identifies as members of the Church of Jesus Christ of Latter-Day Saints (LDS) and patterns of lifestyle variables such as tobacco use differ between LDS and non-LDS individuals,14 we included LDS affiliation as a variable in propensity score models. Because LDS affiliation has a strong negative correlation with tobacco use, LDS affiliation was not included in the main multivariable models to avoid collinearity with ever smoking.
All analyses were conducted using R Statistical Software (v4.3.1; R core team 2023; Vienna, Austria).
Results
Study participants
Cases tended to have a higher education level, were less likely to smoke, and were more likely to engage in physical activity in the past 30 days than controls (Table 1). Men tended to be older at the time of diagnosis than women. BMI was balanced between cases and controls in men. Among women, a high proportion of controls did not provide their weight resulting in missing values for 12%. Although this reduced the ability to compare patterns between cases and controls, a higher proportion of cases among women had a BMI <25 compared with controls. The proportion of men and women identifying as LDS was similar between cases and controls.
Table 1.
Demographic characteristics of in situ and invasive melanoma cases and controls, by sex.
| Men |
Women |
|||||
|---|---|---|---|---|---|---|
| Controls | In situ cases | Invasive cases | Controls | In situ cases | Invasive cases | |
| (N = 3202) | (n = 320) | (n = 321) | (N = 2633) | (n = 246) | (n = 280) | |
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | |
| Age (years) | ||||||
| <45 | 464 (14%) | 44 (14%) | 49 (15%) | 689 (26%) | 65 (26%) | 72 (26%) |
| 45-54 | 349 (11%) | 36 (11%) | 33 (10%) | 477 (18%) | 42 (17%) | 55 (20%) |
| 55-64 | 826 (26%) | 83 (26%) | 82 (26%) | 707 (27%) | 77 (31%) | 64 (23%) |
| 65-74 | 1213 (38%) | 125 (39%) | 118 (37%) | 615 (23%) | 47 (19%) | 76 (27%) |
| 75+ | 350 (11%) | 32 (10%) | 39 (12%) | 145 (6%) | 15 (6%) | 13 (5%) |
| Race and ethnicity | ||||||
| Hispanic | 20 (1%) | * (<5%) | * (<5%) | 30 (1%) | * (<5%) | * (<5%) |
| Non-Hispanic American Indian/Alaskan Native | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Non-Hispanic Asian | * (<1%) | * (<1%) | 0 (0%) | 15 (1%) | * (0%) | * (<5%) |
| Non-Hispanic Black | * (<1%) | 0 (0%) | * (<1%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Non-Hispanic Pacific Islander | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Non-Hispanic White | 3110 (97%) | 311 (97%) | 311 (97%) | 2545 (97%) | 236 (96%) | 273 (98%) |
| Non-Hispanic multiracial | 42 (1%) | * (<5%) | * (<5%) | 33 (1%) | * (<5%) | * (<5%) |
| Non-Hispanic other | * (<1%) | 0 (0%) | * (<5%) | * (<1%) | 0 (0%) | 0 (0%) |
| Missing | 15 (<1%) | * (<1%) | * (<5%) | * (<1%) | * (0%) | 0 (0%) |
| Education | ||||||
| Less than high school diploma | * (<5%) | * (1%) | * (<5%) | * (<5%) | * (<1%) | * (<1%) |
| High school diploma/GED | 563 (18%) | 43 (13%) | 50 (16%) | 545 (21%) | 39 (16%) | 60 (21%) |
| Some college | 981 (31%) | 81 (25%) | 87 (27%) | 989 (38%) | 72 (29%) | 91 (32%) |
| College graduate or more | 1561 (49%) | 191 (60%) | 180 (56%) | 1030 (39%) | 134 (54%) | 127 (45%) |
| Missing | * (<1%) | * (<1%) | * (<1%) | * (<1%) | * (<1%) | * (<1%) |
| Ever smoking | ||||||
| Yes | 975 (30%) | * (26%) | 88 (27%) | * (<25%) | * (19%) | * (<15%) |
| No | 2213 (69%) | 235 (73%) | 233 (73%) | 2076 (79%) | 198 (80%) | 239 (85%) |
| Missing | 14 (<1%) | * (<1%) | 0 (0%) | * (<1%) | * (<1%) | * (<1%) |
| Physical activity in past 30 days | ||||||
| Yes | 2693 (84%) | 283 (88%) | 289 (90%) | 2150 (82%) | 224 (91%) | 254 (91%) |
| No | * (16%) | 37 (12%) | * (<15%) | * (18%) | 22 (9%) | * (<10%) |
| Missing | * (<1%) | 0 (0%) | * (<1%) | * (<1%) | 0 (0%) | * (<1%) |
| Body mass index (kg/m2) | ||||||
| <25 | 726 (23%) | * (24%) | * (<25%) | 851 (32%) | 123 (50%) | * (41%) |
| 25+ | 2404 (75%) | 240 (75%) | 243 (76%) | 1476 (56%) | * (47%) | 154 (55%) |
| Missing | 72 (<5%) | * (<5%) | * (<5%) | 306 (12%) | * (<5%) | * (<5%) |
| Church of Jesus Christ of Latter-Day Saints (LDS) Religion | ||||||
| Yes | 2041 (64%) | 210 (66%) | 192 (60%) | 1640 (62%) | 145 (59%) | 171 (61%) |
| No | 1126 (35%) | * (32%) | * (<40%) | 957 (36%) | * (40%) | * (<40%) |
| Missing | 35 (<5%) | * (<5%) | * (<5%) | 36 (<5%) | * (<5%) | * (<5%) |
Note: Percentages with a less than symbol are not precise to prevent identification of study participants and may cause columns to not sum to 100%. Race and ethnicity groups have been collapsed in this table to further protect study participant anonymity.
Censored due to cell values <11.
Patterns of associations between tattooing exposures and melanoma risk overall
The prevalence of tattooing was 12% among cases and 15% among controls (Table 2). Ever vs never receiving a tattoo was not associated with melanoma risk (melanoma overall, odds ratio [OR] = 0.92, 95% confidence interval [CI] = 0.74 to 1.13; invasive, OR = 0.81, 95% CI = 0.60 to 1.09). Patterns differed by amount of tattoo exposure. Melanoma risk was decreased among those who received 4 or more tattoo sessions (melanoma overall, OR = 0.44, 95% CI = 0.27 to 0.67; invasive, OR = 0.43, 95% CI = 0.24 to 0.80) and among those with 3 or more large tattoos (melanoma overall, OR = 0.26, 95% CI = 0.10 to 0.54; invasive, OR = 0.23, 95% CI = 0.07 to 0.75) compared with never receiving a tattoo. Receiving only 1 tattoo session was associated with an increased risk of overall melanoma (OR = 1.53, 95% CI = 1.16 to 2.00) but not strongly associated with invasive melanoma (OR = 1.25, 95% CI = 0.85 to 1.83), compared with never receiving a tattoo. Receiving a first tattoo before age 20 was associated with a decreased risk of melanoma (overall, OR = 0.74, 95% CI = 0.51 to 1.05), particularly for invasive melanoma (OR = 0.48, 95% CI = 0.29 to 0.82) compared with never getting tattooed.
Table 2.
Tattooing exposures and risk of overall and invasive melanoma in individuals ages 19-79 years.
| Controls (N = 5835) |
Overall (n = 1167) |
Invasive (n = 601) |
|||
|---|---|---|---|---|---|
| n (%) | n (%) | OR (95% CI) | n (%) | OR (95% CI) | |
| Ever tattooed | |||||
| No | 4980 (85%) | 1027 (88%) | Referent | 536 (89%) | Referent |
| Yes | 855 (15%) | 140 (12%) | 0.92 (0.74 to 1.13) | 65 (11%) | 0.81 (0.60 to 1.09) |
| Time since first tattoo | |||||
| No tattoo sessions | 4980 (85%) | 1027 (88%) | Referent | 536 (89%) | Referent |
| <10 years | 198 (3%) | 41 (4%) | 1.11 (0.76 to 1.58) | 17 (3%) | 0.87 (0.52 to 1.47) |
| 10+ years | 649 (11%) | 96 (8%) | 0.83 (0.64 to 1.06) | 45 (7%) | 0.74 (0.52 to 1.04) |
| Age at first tattoo | |||||
| No tattoo sessions | 4980 (85%) | 1027 (88%) | Referent | 536 (89%) | Referent |
| <20 | 317 (5%) | 42 (4%) | 0.74 (0.51 to 1.05) | 17 (3%) | 0.48 (0.29 to 0.82) |
| 20+ | 530 (9%) | 95 (8%) | 0.98 (0.76 to 1.25) | 45 (7%) | 0.83 (0.60 to 1.14) |
| No. of tattoo sessions | |||||
| No tattoo sessions | 4980 (85%) | 1027 (88%) | Referent | 536 (89%) | Referent |
| 1 | 280 (5%) | 79 (7%) | 1.53 (1.16 to 2.00) | 34 (6%) | 1.25 (0.85 to 1.83) |
| 2-3 | 277 (5%) | 37 (3%) | 0.73 (0.50 to 1.04) | 19 (3%) | 0.72 (0.44 to 1.17) |
| 4 or more | 298 (5%) | 24 (2%) | 0.44 (0.27 to 0.67) | 12 (2%) | 0.43 (0.24 to 0.80) |
| No. of large tattoos | |||||
| No tattoo sessions | 4980 (85%) | 1027 (88%) | Referent | 536 (89%) | Referent |
| 0 | 407 (7%) | 80 (7%) | 1.06 (0.81 to 1.38) | 37 (6%) | 0.93 (0.64 to 1.34) |
| 1-2 | 308 (5%) | * (<5%) | 0.93 (0.66 to 1.28) | * (<5%) | 0.85 (0.54 to 1.33) |
| 3 or more | 140 (2%) | * (<5%) | 0.26 (0.10 to 0.54) | * (<1%) | 0.23 (0.07 to 0.75) |
Note: Percentages with a less than symbol are not precise to prevent identification of study participants and may cause columns to not sum to 100%. Abbreviations: CI = confidence interval; OR = odds ratio.
Adjusted for age (5-year groups), sex, race and ethnicity (non-Hispanic White, Hispanic, all other racial and ethnic groups), education (<high school diploma, high school diploma, some college, 4-year college degree or more), ever smoking (yes/no), physical activity in the past 30 days (yes/no), and body mass index (<25/25+ kg/m2).
Censored to hide cell values <11.
Patterns of associations between tattooing exposures and melanoma risk stratified by sex
The prevalence of tattooing was lower among men (8% among cases and 12% among controls) compared with women (17% among cases and 18% among controls; Table 3). Ever receiving a tattoo was associated with a decreased risk of melanoma among men (overall melanoma, OR = 0.74, 95% CI = 0.53 to 1.02), but not among women (overall melanoma, OR = 1.12, 95% CI = 0.84 to 1.48). The decreased melanoma risk associated with 4 or more tattoo sessions was stronger among men (overall melanoma, OR = 0.25, 95% CI = 0.09 to 0.53; invasive, OR = 0.15, 95% CI = 0.04 to 0.62), and less pronounced among women (overall melanoma, OR = 0.64, 95% CI = 0.36 to 1.07; invasive, OR = 0.74, 95% CI = 0.37 to 1.48). The increased melanoma risk associated with 1 tattoo session was much stronger among women (overall melanoma, OR = 1.90, 95% CI = 1.30 to 2.74; invasive, OR = 1.41, 95% CI = 0.83 to 2.39) than men (overall melanoma, OR = 1.21, 95% CI = 0.79 to 1.81; invasive, OR = 1.12, 95% CI = 0.64 to 1.97), with invasive melanoma risk among men close to null (OR = 1.12, 95% CI = 0.64 to 1.97). Receiving a first tattoo before age 20 was associated with decreased melanoma risk among men (overall melanoma, OR = 0.55, 95% CI = 0.31 to 0.90; invasive, OR = 0.49, 95% CI = 0.23 to 1.03). Among women, receiving a first tattoo before age 20 was associated with decreased risk of invasive (OR = 0.64, 95% CI = 0.29 to 1.42), but not overall melanoma (OR = 1.06, 95% CI = 0.63 to 1.74).
Table 3.
Tattooing exposures and risk of overall and invasive melanoma in men and women ages 19-79 years.
| Men |
Women |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Controls (N = 5835) n (%) |
Cases |
Cases |
||||||||
| Overall (n = 641) |
Invasive (n = 321) |
Controls (N = 2633) |
Overall (n = 526) |
Invasive (n = 280) |
||||||
| n (%) | OR (95% CI) | (%) | OR (95% CI) | n (%) | n (%) | OR (95% CI) | (%) | OR (95% CI) | ||
| Ever tattooed | ||||||||||
| No | 2832 (88%) | 589 (92%) | Referent | (92%) | Referent | 2148 (82%) | 438 (83%) | Referent | (86%) | Referent |
| Yes | 370 (12%) | 52 (8%) | 0.74 (0.53 to 1.02) | (8%) | 0.70 (0.45 to 1.10) | 485 (18%) | 88 (17%) | 1.12 (0.84 to 1.48) | (14%) | 0.97 (0.65 to 1.43) |
| Time since first tattoo | ||||||||||
| No tattoo sessions | 2832 (88%) | 589 (92%) | Referent | (92%) | Referent | 2148 (82%) | 438 (83%) | Referent | (86%) | Referent |
| <10 years | 64 (2%) | 11 (2%) | 0.90 (0.43 to 1.69) | (<5%) | 0.60 (0.21 to 1.70) | 134 (5%) | 30 (6%) | 1.26 (0.80 to 1.94) | (5%) | 1.06 (0.57to 1.95) |
| 10+ years | 303 (9%) | 41 (6%) | 0.71 (0.49 to 1.02) | (7%) | 0.73 (0.45 to 1.19) | 346 (13%) | 55 (10%) | 0.99 (0.69 to 1.38) | (8%) | 0.80 (0.49 to 1.30) |
| Age at first tattoo | ||||||||||
| No tattoo sessions | 2832 (88%) | 589 (92%) | Referent | (92%) | Referent | 2148 (82%) | 438 (83%) | Referent | (86%) | Referent |
| <20 | 169 (5%) | 19 (<5%) | 0.55 (0.31 to 0.90) | (<5%) | 0.49 (0.23 to 1.03) | 148 (6%) | 23 (4%) | 1.06 (0.63 to 1.74) | (<5%) | 0.64 (0.29 to 1.42) |
| 20+ | 198 (6%) | 33 (5%) | 0.90 (0.60 to 1.32) | (5%) | 0.90 (0.53 to 1.51) | 332 (13%) | 62 (12%) | 1.08 (0.78 to 1.47) | (10%) | 0.96 (0.62 to 1.49) |
| No. of tattoo sessions | ||||||||||
| No tattoo sessions | 2832 (88%) | 589 (92%) | Referent | (92%) | Referent | 2148 (82%) | 438 (83%) | Referent | (86%) | Referent |
| 1 | 140 (4%) | 32 (5%) | 1.21 (0.79 to 1.81) | (5%) | 1.12 (0.64 to 1.97) | 140 (5%) | 47 (9%) | 1.90 (1.30 to 2.74) | (6%) | 1.41 (0.83 to 2.39) |
| 2-3 | 103 (3%) | * (<5%) | 0.67 (0.35 to 1.18) | (<5%) | 0.80 (0.38 to 1.71) | 174 (7%) | 24 (5%) | 0.82 (0.50 to 1.29) | (<5%) | 0.75 (0.39 to 1.44) |
| 4 or more | 127 (4%) | * (<5%) | 0.25 (0.09 to 0.53) | (<5%) | 0.15 (0.04 to 0.62) | 171 (6%) | 17 (3%) | 0.64 (0.36 to 1.07) | (<5%) | 0.74 (0.37 to 1.48) |
| No. of large tattoos | ||||||||||
| No tattoo sessions | 2832 (88%) | 589 (92%) | Referent | (92%) | Referent | 2148 (82%) | 438 (83%) | Referent | (86%) | Referent |
| 0 | 148 (5%) | * (<5%) | 0.82 (0.50 to 1.29) | (<5%) | 0.86 (0.46 to 1.59) | 259 (10%) | 57 (11%) | 1.27 (0.90 to 1.76) | (9%) | 1.05 (0.66 to 1.67) |
| 1-2 | 143 (4%) | 26 (4%) | 0.99 (0.62 to 1.53) | (<5%) | 0.94 (0.51 to 1.74) | 165 (6%) | * (<10%) | 0.91 (0.55 to 1.44) | (<5%) | 0.82 (0.42 to 1.58) |
| 3 or more | 79 (2%) | * (<1%) | 0.07 (0.00 to 0.30) | (<1%) | – | 61 (2%) | * (<5%) | 0.59 (0.20 to 1.41) | (<5%) | 0.66 (0.20 to 2.20) |
Note: Percentages with a less than symbol are not precise to prevent identification of study participants and may cause columns to not sum to 100%.
Adjusted for age (5-year groups), race and ethnicity (non-Hispanic White, Hispanic, all other racial and ethnic groups), education (<high school diploma, high school diploma, some college, 4-year college degree or more), ever smoking (yes/no), physical activity in the past 30 days (yes/no), and body mass index (<25/25+ kg/m2).
Empty cells indicate no estimate could be made due to the sample size.
Censored to hide cell values <11.
Patterns of associations between tattooing exposures and in situ melanoma
Ever receiving a tattoo was not associated with risk of in situ melanoma (OR = 1.04, 95% CI = 0.78 to 1.39; Table S1). Receiving 1 tattoo session was associated with increased risk of in situ melanoma (OR = 1.85, 95% CI = 1.31 to 2.63), particularly among women (OR = 2.47, 95% CI = 1.55 to 3.93). As with invasive melanoma, risk of in situ melanoma was decreased with 4 or more tattoo sessions (OR = 0.44, 95% CI = 0.24 to 0.83) and 3 or more large tattoos (OR = 0.28, 95% CI = 0.09 to 0.89). Receiving a first tattoo before age 20 was associated with decreased in situ melanoma risk among men (OR = 0.63, 95% CI = 0.31 to 1.27) but increased in situ melanoma risk among women (OR = 1.58, 95% CI = 0.84 to 2.98).
Distribution of melanoma risk factors among cases and assessment of unmeasured confounding
Behaviors associated with increased melanoma risk tended to be associated with being tattooed among cases. In particular, a higher proportion of tattooed cases had ever used an indoor tanning bed; <40% of men and 69% of women with 2 or more tattoo sessions, and <30% of men and 53% of women with 1 tattoo session reported using a tanning device 10 or more times compared with 19% of men and 39% of women with no tattoo sessions (Table S2; percentages based on small numbers are shown with a less than symbol to prevent identification of study participants). The proportion of men and women who reported at least 1 red or painful sunburn in the 1- to 2-year period before diagnosis was highest among those with 1 tattoo session (53% among men, 51% among women) compared with those with no tattoo sessions (41% among men, 38% among women) and those with 2 or more sessions (<40% among men, 47% among women).
We did not see any major differences between frequency-matched and propensity score-matched results by number of tattoo sessions (Table S3). Among in situ cases, tattooed individuals were more likely to have a lesion on the trunk (32%) and less likely to have a lesion on the head and neck (23%) compared with never tattooed individuals (23% and 35%, respectively; Table S4). Among invasive cases, tattooed individuals were more likely to have a lesion on the lower limbs (25%) and less likely to have a lesion on the upper limbs (18%) compared with never tattooed individuals (16% and 27%, respectively). We did not observe that melanomas were colocated with tattoos; less than 7% of tattooed melanoma cases reported that they had a melanoma lesion within 6 inches of a tattoo.
In sensitivity analyses restricted to non-Hispanic White individuals (Table S5), results were generally similar to the main findings, with slightly lower point estimates in some comparisons; for example, the odds ratio for ever having a tattoo and risk of invasive melanoma was 0.81 (95% CI = 0.60 to 1.09) in the main analysis and 0.75 (95% CI = 0.66 to 1.03) in this subgroup.
Discussion
Although the results of this study suggest potentially important roles for tattooing in the incidence of melanoma, as one of the first epidemiological studies of tattooing and melanoma risk, the findings should be considered as generating hypotheses for further study. Decreased melanoma risk was observed among individuals with higher levels of tattoo exposure, both for numbers of sessions and numbers of large tattoos, particularly for invasive melanoma. These patterns do not support previously hypothesized associations between tattoos and increased melanoma risk based on case reports. We did observe increased risk within the lowest level of exposure (1 tattoo session), primarily restricted to in situ melanoma, which may suggest overdiagnosis. It is possible that individuals with tattoos, particularly those receiving only one, are more likely to examine their skin and identify potentially concerning lesions than those who are never tattooed.
The biological mechanisms underlying the observed associations are not known. We initially hypothesized that tattooing may be associated with increased melanoma risk through exposure to carcinogens and inflammatory and immune responses, but the patterns of decreasing risk associated with higher levels of tattoo exposure do not support this hypothesis. It is possible that tattooing may elicit beneficial immune responses that could contribute to immune surveillance of precancerous cells, preventing progression to melanoma.24 Additionally, large tattoos in sun-exposed skin could block ultraviolet (UV) radiation, a major risk factor for melanoma. A prior mouse study observed delayed development of UV-induced squamous cell carcinoma among mice tattooed in black ink compared with control mice who were “tattooed” without ink, with the study’s authors hypothesizing this was due to black pigments absorbing UV radiation and resulting in decreased backscattered radiation exposure.25 We lacked detailed information on tattoo location and were unable to explore this hypothesis.
There is a high likelihood of unmeasured confounding in this study because we did not have data on many melanoma risk factors among controls, in particular the known melanoma risk factors tanning bed use, sun exposure, family history of melanoma, ability to tan/sunburn, prior sunburns, and hair and eye color, although we did have this information on cases. In our study, we observed that melanoma cases with tattoos were more likely to use tanning beds, and that ever use of tanning beds increased with higher number of tattoo sessions. We initially anticipated this could be an issue if we observed consistent increased risks of melanoma associated with tattooing. That is, if we observed an increased risk of melanoma associated with tattooing, it could be caused by higher tanning device use among tattooed compared with nontattooed individuals. Unmeasured confounding by tanning bed use may attenuate the decreased risks we observed. It is possible that the increased risk of melanoma that we observed with 1 tattoo session could be due to unmeasured confounding because a higher proportion of cases with 1 tattoo session reported at least 1 sunburn in the 1-2 years before melanoma diagnosis compared with those who were never tattooed or those with 2 or more tattoo sessions.
It is also possible that people with tattoos have fewer melanoma risk factors than those without tattoos. Although the relationships between melanoma risk factors and tattoos are not fully characterized, a prior study among cancer-free individuals reported that tattooing was associated with a history of painful sunburns and ever use of tanning devices, in addition to a lower likelihood of having blue eyes.26 In another study, a higher proportion of tattooed individuals reported both heavier sun exposure and greater use of sun protection compared with nontattooed individuals.27 They also observed individuals with multiple tattoos used sunscreen more frequently and used sunscreens with a higher sun protection factor compared with individuals with 1 tattoo, who were more likely to use long sleeves and parasols as sun protection than those with multiple tattoos.27 A recent study of individuals surveyed at beaches in Poland observed that tattooed individuals were more likely to report never using sunscreen than nontattooed individuals (20% and 13%, respectively).28 However, among individuals who reported using sunscreen, tattooed individuals were more likely to use sunscreens with a higher sun protection factor than nontattooed individuals.28 These findings suggest there may be distinct subgroups of people with tattoos—1 with riskier sun exposure behaviors and another with more protective behaviors. As patterns of sun exposure and tattooing are variable and were not collected in our data, the impact of these relationships on our results is not known. It is plausible that the decreased risks of melanoma observed with tattooing were mediated by sun protective behaviors.27,28 It is also possible that people with certain underlying melanoma risk factors such as a large number of moles would be less likely to receive a tattoo, although to our knowledge there are no published data to support or refute this. The DAG that we present in Figure S1 shows that in addition to potential unmeasured confounding, post-tattoo UV exposure may be a key potential mediator in this study. Patterns of UV exposure among people with tattoos may differ from people without tattoos for a number of different reasons including avoiding the sun to prevent tattoos from fading, or increased sun exposure due to increased risk-taking tendencies among individuals with tattoos or a desire for tattoos to be visible.
Smoking may also be a confounder because tattooing is strongly associated with tobacco smoking14,15,29,30 and prior research has suggested smoking is associated with a decreased risk of melanoma.31 Although we adjusted for ever smoking, it is possible that there was residual confounding.
We examined the potential effects of unmeasured confounding through propensity score analyses to examine possible confounders in the relationships between ever tattooing and number of tattoo sessions with melanoma risk. The similarities between the estimates from the original models and the estimates from the propensity score-matched models suggest that unmeasured confounding by factors associated with the available health and demographic variables is not likely to be the cause of the associations we observed. Still, we are unable to rule out the possibility of unmeasured confounding because of the absence of melanoma risk variables among controls. The relationships between tattooing and sun exposure behaviors including both increased and decreased sun protection reported in prior studies27,28 highlight the need for future studies to collect detailed data on all aspects of sun/UV exposure.
The existing literature on tattooing and skin cancer risk is extremely limited. A recent study from Clemmensen et al. composed of 2 twin-based studies in Denmark reported increased risk of combined melanoma and nonmelanoma skin cancers associated with ever receiving a tattoo (case-cotwin hazard ratio [HR] = 1.62, 95% CI = 1.08 to 2.41; cohort HR = 3.91, 95% CI = 1.42 to 10.8) and ever receiving a large tattoo (case-cotwin HR = 2.37, 95% CI = 1.11 to 5.06).12 Within their cohort study, they also reported increased risk of basal cell carcinoma associated with ever tattooing (HR = 2.83, 95% CI = 1.42 to 10.78).12 The study did not report number of tattoo sessions, but if a large proportion of participants had only 1 tattoo session, our findings would be consistent with theirs. Clemensen et al.’s study of all skin cancers combined included only a small number of individuals with tattoos (8-30 tattooed cancer cases depending on the study design), and they controlled for smoking, physical exercise, alcohol consumption, and education but no other melanoma risk factors.12 A recent case-control study in Sweden by Liljedahl et al. observed no association between tattoos and squamous cell carcinoma of the skin.32 Another study by Barton et al. observed a decreased prevalence of tattooing among individuals with early-onset basal cell carcinoma compared with cancer-free controls.26 In a letter to the editor that presented a reanalysis of the data, there was a 40% decreased risk of basal cell cancer among individuals who were ever tattooed (crude OR = 0.6, 95% CI = 0.5 to 0.7).33 Barton et al. also reported an 80% increased risk of basal cell carcinomas co-occurring in the region of the body that was tattooed.26
Strengths and limitations
The population-based study design using Surveillance, Epidemiology, and End Results registry data was a strength because it ensured complete ascertainment of all histologically confirmed melanoma cases in Utah during the study period. Further, the use of BRFSS data ensured that controls were reflective of the underlying population from which cases arose. Because controls were selected from the BRFSS, we were limited to the variables included on the BRFSS, and a key limitation of this study was the lack of data on potential confounders among controls. Despite our attempts to estimate the potential impact of unmeasured confounding on our study, we are unable to rule out the possibility that our results could be affected by the inability to account for key melanoma risk factors or residual confounding by smoking because we were only able to control for ever smoking. We also were not able to ask controls additional questions on tattooing that may be relevant to melanoma risk including tattoo sun exposure. We were further limited by the 41% response proportion among cases, although this response proportion is typical for interview-based studies.34 Although study recruitment materials described the study aims broadly in terms of lifestyle exposures rather than specifically emphasizing tattoos to reduce the likelihood of differential participation based on tattoo status, and participants and nonparticipants were similar with respect to sex, age, and race, it is possible that nonparticipants had a different prevalence of tattooing than participants. Despite these limitations, this study provides some of the first evidence to begin understanding how tattooing may relate to melanoma risk and provides justification for future studies that should collect data on key potential confounders.
Conclusions
Our findings suggest higher tattoo exposure such as more tattoo sessions or more large tattoos may be associated with decreased melanoma risk. The biological mechanisms underlying these associations are not known but could be related to tumor immune surveillance or differences in melanoma risk factors between those with and without tattoos. It is also possible that sun exposure-related behaviors associated with tattooing influenced our results. Future larger studies with detailed data on potential confounders are needed to clarify the relationships between tattooing and melanoma risk.
Supplementary Material
Acknowledgments
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funders of this work did not play a role in the design of this study; the collection, analysis, or interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication. Where authors are identified as personnel of the International Agency for Research on Cancer/World Health Organization, the authors alone are responsible for the views expressed in this article, and they do not necessarily represent the decisions, policy, or views of the International Agency for Research on Cancer/World Health Organization. This work was previously presented in part as a poster at the 2024 American Association for Cancer Research annual meeting and as an oral presentation at the 2025 World Congress of Tattoo and Pigment Research. We wish to thank the participants in this study who dedicated their time to make this research possible. We also thank Lori Burke, Carrie Bateman, and Kacey Wigren at the Utah Cancer Registry, whose work contributed to this study.
Contributor Information
Rachel D McCarty, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; University of Utah Intermountain Healthcare Department of Population Health Sciences Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States; Environment and Lifestyle Epidemiology Branch, International Agency for Research on Cancer, Lyon, France.
Britton Trabert, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; University of Utah Intermountain Healthcare Department of Population Health Sciences Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States; Department of Obstetrics and Gynecology, Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States.
Lindsay J Collin, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; University of Utah Intermountain Healthcare Department of Population Health Sciences Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States; Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA, United States.
Morgan M Millar, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; Department of Internal Medicine, University of Utah, Salt Lake City, UT, United States; Utah Cancer Registry, University of Utah, Salt Lake City, UT, United States.
David Kriebel, Lowell Center for Sustainable Production, University of Massachusetts Lowell, Lowell, MA, United States.
Laurie Grieshober, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; University of Utah Intermountain Healthcare Department of Population Health Sciences Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States.
Mollie E Barnard, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; University of Utah Intermountain Healthcare Department of Population Health Sciences Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States; Slone Epidemiology Center, Boston University Chobanian and Avedisian School of Medicine, Boston, MA, United States.
Jenna Sawatzki, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States.
Marjorie Carter, Utah Cancer Registry, University of Utah, Salt Lake City, UT, United States.
Valerie Yoder, Utah Cancer Registry, University of Utah, Salt Lake City, UT, United States.
Jeffrey A Gilreath, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States.
Douglas Grossman, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; Department of Dermatology, University of Utah Health Sciences Center, Salt Lake City, UT, United States.
John Hyngstrom, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; Department of Surgery, Rush University, Chicago, IL, United States.
Paul J Shami, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; Department of Internal Medicine, Division of Hematology and Hematologic Malignancies, University of Utah, Salt Lake City, UT, United States.
Jennifer A Doherty, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, United States; University of Utah Intermountain Healthcare Department of Population Health Sciences Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, UT, United States; Utah Cancer Registry, University of Utah, Salt Lake City, UT, United States.
Author Contributions
Rachel D. McCarty (Conceptualization, Formal analysis, Investigation, Methodology, Writing—original draft), Britton Trabert (Investigation, Methodology, Writing—review & editing), Lindsay J. Collin (Investigation, Methodology, Writing—review & editing), Morgan M. Millar (Investigation, Methodology, Writing—review & editing), David Kriebel (Investigation, Methodology, Supervision, Writing—review & editing), Laurie Grieshober (Investigation, Methodology, Writing—review & editing), Mollie E. Barnard (Investigation, Methodology, Writing—review & editing), Jenna Sawatzki (Investigation, Methodology, Writing—review & editing), Marjorie Carter (Investigation, Methodology, Writing—review & editing), Valerie Yoder (Investigation, Writing—review & editing), Jeffrey A. Gilreath (Investigation, Methodology, Writing—review & editing), Douglas Grossman (Methodology, Writing—review & editing), John Hyngstrom (Methodology, Writing—review & editing), Paul J. Shami (Methodology, Writing—review & editing), and Jennifer A. Doherty (Funding acquisition, Investigation, Methodology, Supervision, Writing—review & editing)
Supplementary material
Supplementary material is available at JNCI: Journal of the National Cancer Institute online.
Funding
We acknowledge the direct financial support for the research reported in this publication provided by the Huntsman Cancer Foundation, the Huntsman Cancer Institute Melanoma Disease Center, and the Cancer Control and Population Sciences (CCPS) Program at Huntsman Cancer Institute. We also acknowledge support by the National Cancer Institute of the National Institutes of Health (NIH) under Award Number P30CA042014. The Utah Cancer Registry is funded by the National Cancer Institute’s SEER Program, Contract No. HHSN261201800016I, the US Centers for Disease Control and Prevention’s National Program of Cancer Registries, Cooperative Agreement No. NU58DP007131, with additional support from the University of Utah and Huntsman Cancer Foundation. Data collection and management via REDCap was supported by UL1TR002538 NCATS/NIH. Rachel McCarty was supported in part by the National Center for Advancing Translational Sciences of the NIH under Award Number T32TR004392. Mollie E. Barnard received support from K00CA212222 from the National Cancer Institute of the NIH. Lindsay J. Collin was supported by R00CA277580 from the National Cancer Institute of the National Institutes of Health.
Conflict of Interest
Institutional pilot funding from the Huntsman Cancer Institute supported this work. Lindsay J. Collin was supported by R00CA277580 from the National Cancer Institute of the National Institutes of Health. Lindsay J. Collin reports personal fees from Epidemiologic Research & Methods, LLC, outside of the submitted work.
Data Availability
The data from cancer cases used in this study were obtained from the Utah Cancer Registry, which collects and maintains confidential data in accordance with Utah law. These data are not publicly available but can be accessed through review and approval of a proposed study protocol by the University of Utah Institutional Review Board and the Resource for Genetic and Epidemiologic Research. Data from Utah BRFSS participants who were used as controls in this study are restricted and maintained by the Utah Department of Health and Human Services, through which these data can be requested.
References
- 1. Schaeffer K, Dinesh S. 32% of Americans have a tattoo, including 22% who have more than one. Pew Research Center. Accessed August 16, 2023. https://www.pewresearch.org/short-reads/2023/08/15/32-of-americans-have-a-tattoo-including-22-who-have-more-than-one/.
- 2. Foerster M, Schreiver I, Luch A, Schüz J. Tattoo inks and cancer. Cancer Epidemiol. 2020;65:101655-101655. [DOI] [PubMed] [Google Scholar]
- 3. Regensburger J, Lehner K, Maisch T, et al. Tattoo inks contain polycyclic aromatic hydrocarbons that additionally generate deleterious singlet oxygen: PAHs in tattoo inks. Exp Dermatol. 2009;19:e275–81. [Google Scholar]
- 4. Engel E, Spannberger A, Vasold R, König B, Landthaler M, Bäumler W. Photochemical cleavage of a tattoo pigment by UVB radiation or natural sunlight. J Dtsch Dermatol Ges 2007;5:583-589. [DOI] [PubMed] [Google Scholar]
- 5. Hauri U, Hohl C. Photostability and breakdown products of pigments currently used in tattoo inks. Curr Probl Dermatol. 2015;48:164-169. [DOI] [PubMed] [Google Scholar]
- 6. Engel E, Vasold R, Santarelli F, et al. Tattooing of skin results in transportation and light-induced decomposition of tattoo pigments—a first quantification in vivo using a mouse model: tattoo pigments in skin. Exp Dermatol. 2010;19:54-60. [DOI] [PubMed] [Google Scholar]
- 7. Lehner K, Santarelli F, Vasold R, et al. Black tattoos entail substantial uptake of genotoxicpolycyclic aromatic hydrocarbons (PAH) in human skin and regional lymph nodes. PLoS One. 2014;9:e92787. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kaur RR, Kirby W, Maibach H. Cutaneous allergic reactions to tattoo ink. J Cosmet Dermatol. 2009;8:295-300. [DOI] [PubMed] [Google Scholar]
- 9. Kluger N. Cutaneous and systemic complications associated with tattooing. Presse Med 2016;45:567-576. [DOI] [PubMed] [Google Scholar]
- 10. Kalaora S, Nagler A, Wargo JA, Samuels Y. Mechanisms of immune activation and regulation: lessons from melanoma. Nat Rev Cancer. 2022;22:195-207. [DOI] [PubMed] [Google Scholar]
- 11. Lebhar J, Jacobs J, Rundle C, Kaplan SJ, Mosca PJ. Systematic review and analysis of published reports of skin cancers arising within tattoos. JAAD Int. 2024;16:133-143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Clemmensen SB, Mengel-From J, Kaprio J, Frederiksen H, von Bornemann Hjelmborg J. Tattoo ink exposure is associated with lymphoma and skin cancers—a Danish study of twins. BMC Public Health. 2025;25:170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Braverman S. One in five U.S. adults now has a tattoo. Accessed June 8, 2022. https://www.prnewswire.com/news-releases/one-in-five-us-adults-now-has-a-tattoo-140123523.html.
- 14. McCarty RD, Trabert B, Millar MM, et al. Associations of demographic, health, and risk-taking behaviors with tattooing in a population-based cross-sectional study of ∼18,000 US adults. 2024; 10.21203/rs.3.rs-4838597/v1 [DOI]
- 15. Morlock R, Morlock A. Think before you ink: perception, prevalence, and correlates of tattooing and tattoo regret in US adults. Cureus. 2023;15:e48167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Olsen CM, Thompson JF, Pandeya N, Whiteman DC. Evaluation of sex-specific incidence of melanoma. JAMA Dermatol. 2020;156:553-560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Adamson AS, Naik G, Jones MA, Bell KJ. Ecological study estimating melanoma overdiagnosis in the USA using the lifetime risk method. BMJ Evid Based Med. 2024;29:156-161. [Google Scholar]
- 18. Didier AJ, Nandwani SV, Watkins D, et al. Patterns and trends in melanoma mortality in the United States, 1999–2020. BMC Cancer. 2024;24:790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Millar MM, Herget K, Howlett C, Codden R, Doherty JA. Cancer in Utah: incidence and mortality statistics through 2021. Utah Cancer Registry, 2024. https://uofuhealth.utah.edu/documents/cancerinutah2024.
- 20. Mounessa JS, Caravaglio JV, Dellavalle RP. Comparison of regional and state differences in melanoma rates in the United States: 2003 vs 2013. JAMA Dermatol. 2017;153:345-347. [DOI] [PubMed] [Google Scholar]
- 21. McCarty RD, Trabert B, Kriebel D, et al. Tattoos and risk of hematologic cancer: a population-based case-control study in Utah. Cancer Med. 2024;13:e70260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Månsson R, Joffe MM, Sun W, Hennessy S. On the estimation and use of propensity scores in case-control and case-cohort studies. Am J Epidemiol. 2007;166:332-339. [DOI] [PubMed] [Google Scholar]
- 23. Ho DE, Imai K, King G, Stuart EA. MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Softw. 2011;42:1-28. [Google Scholar]
- 24. Marzagalli M, Ebelt ND, Manuel ER. Unraveling the crosstalk between melanoma and immune cells in the tumor microenvironment. Semin Cancer Biol. 2019;59:236-250. [DOI] [PubMed] [Google Scholar]
- 25. Lerche CM, Sepehri M, Serup J, Poulsen T, Wulf HC. Black tattoos protect against UVR-induced skin cancer in mice. Photodermatol Photoimmunol Photomed. 2015;31:261-268. [DOI] [PubMed] [Google Scholar]
- 26. Barton DT, Zens MS, Marmarelis EL, Gilbert-Diamond D, Karagas MR. Cosmetic tattooing and early onset basal cell carcinoma: a population-based case-control study from New Hampshire. Epidemiol Camb Mass. 2020;31:448-450. [Google Scholar]
- 27. Kluger N, Shourick J, Seité S, Taieb C. Sun protection and sun exposure habits among tattooed individuals. JEADV Clin Pract. 2024;3:1318-1324. [Google Scholar]
- 28. Rogowska P, Nette D, Bulinska N, Sobjanek M, Slawinska M. Tattoos and sun protection: do beachgoers with body ink take more risks? In: 7th World Congress of Tattoo and Pigment Research, 2025, 61.
- 29. Karagas MR, Wasson JH. A World Wide Web-based survey of nonmedical tattooing in the United States. J Am Acad Dermatol. 2012;66:e13-14-e14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Mortensen K, French MT, Timming AR. Are tattoos associated with negative health-related outcomes and risky behaviors? Int J Dermatol. 2019;58:816-824. [DOI] [PubMed] [Google Scholar]
- 31. Friedman EB, Williams GJ, Lo SN, Thompson JF. Effect of smoking on melanoma incidence: a systematic review with meta-analysis. J Natl Cancer Inst. 2024;116:1739-1752. [DOI] [PubMed] [Google Scholar]
- 32. Liljedahl ER, Engfeldt M, Nielsen K, Jöud A, Nielsen C. Tattoos and cutaneous squamous cell carcinoma: a population-based case-control study. Eur J Epidemiol. 2025;40:451-461., published online April 25. 10.1007/s10654-025-01230-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Kluger N, Dub T. Re: “Cosmetic tattooing and early onset basal cell carcinoma: a population-based case control study from New Hampshire.” Epidemiology. 2020;31:e39-e39. [Google Scholar]
- 34. Galea S, Tracy M. Participation rates in epidemiologic studies. Ann Epidemiol. 2007;17:643-653. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data from cancer cases used in this study were obtained from the Utah Cancer Registry, which collects and maintains confidential data in accordance with Utah law. These data are not publicly available but can be accessed through review and approval of a proposed study protocol by the University of Utah Institutional Review Board and the Resource for Genetic and Epidemiologic Research. Data from Utah BRFSS participants who were used as controls in this study are restricted and maintained by the Utah Department of Health and Human Services, through which these data can be requested.