Point prevalence of non-melanoma and melanoma skin cancers in Australian surfers and swimmers in Southeast Queensland and Northern New South Wales

Study design

This was a cross-sectional study design incorporating a survey followed by whole-body skin cancer screening. Participants were recruited from notices sent to local general practitioners (GPs), surfing and swimming clubs in addition to local media coverage.

Survey

The survey consisted of four sections which included participants’ physiological demographics (“Introduction”), activity-specific demographics (peak UVR exposure; “Materials and Methods”) (NSW Government: SafeWork, 2020; Gonzaga, 2009) and history of skin cancer (Results). “Discussion” was completed by a clinician following the screening and included Fitzpatrick skin type (Gupta & Sharma, 2019), type(s) and location(s) of skin cancer(s). “Conclusions” detailed histopathology results. All surveys were completed within the clinic waiting room. If participants had any questions pertaining to the survey, they were answered by either one of the researchers who was onsite or the specialist, depending upon the nature of the query.

Screening strategy

Whole body skin checks were conducted using a handheld dermatoscope, completed by an accredited skin cancer doctor focusing upon AK, NMSC and MSC using the chaos and clues algorithm (Rosendahl et al., 2012) and prediction without pigment, a decision algorithm for non-pigmented skin malignancy (Rosendahl et al., 2014). These methods have been shown to significantly increase the diagnostic accuracy of detecting AK, NMSC and MSC when used by experienced clinicians (Dinnes et al., 2018; Dinnes et al., 2018). A dermatoscope with 10× magnification and LEDHQ illumination (Heine Delta 30; Heine, Optotechnic GhbH, Herrsching, Germany) was used to inspect all suspect skin lesions. Confirmation of all skin cancers was attained via histopathology from a commercial laboratory.

Statistical analysis

Normality of data was assessed via kurtosis, skewness, Q–Q plots and the Kolmogorov–Smirnov test (with Lilliefors significance correction). Heteroscedasticity was assessed with Levene’s inferential test. Statistical analyses were completed using IBM’s Statistical Package for Social Sciences (SPSS, Ver. 27.0), Excel (Microsoft Office 365; Microsoft Corporation, Redmond, WA, USA) and included demographics, independent sample t-tests, Chi-square tests and ANOVA (Bonferroni post-hoc test) were used to determine significance between the number of skin cancers and decade of age groups. A bivariate (two-tailed) Pearson correlation coefficient was used to determine relationships between selected outcome variables (number of skin cancers to age and body mass index). Alpha was set a priori at P < 0.05 to determine significance between groups.

Point prevalence was determined by the number of surfers or swimmers with an AK, BCC, SCC, SCC in situ or MSC divided by the total number in that group (National Institute of Mental Health, 2022). The standardized rate per 100,000 was calculated as, for example, the number of surfers or swimmers identified with a particular skin cancer multiplied by 100,000, then divided by the total number of surfers or swimmers, respectively (Statistics Canada, 2017). Odds ratios (OR) is a ratio of two sets of odds, the odds of the event occurring in one group (i.e. surfers) vs. the odds of the event occurring in a different group (i.e. swimmers). The OR of surfers to swimmers was then determined by the odds in surfers divided by the odds in swimmers (Tenny & Hoffman, 2022). MedCalc statistical software (https://www.medcalc.org/calc/odds_ratio.php) was used to calculate all odds ratios. Comparative standardized rates for AK, BCC, SCC and melanoma for the general Australia population were identified from the literature (Lai, Cranwell & Sinclair, 2018; Australian Institute of Health and Welfare, 2020; Smith, 2019).

Sample characteristics/demographics

A total of 171 participants (males n = 94, females n = 77; surfers n = 116, swimmers n = 55) competed the survey and underwent a total body skin check for AK, BCC, SCC in situ, SCC, and MSCs. Swimmers and surfers had similar mean age, height, mass, body mass index (BMI) and body surface area (BSA). When groups were combined, there was a significant correlation between the total number of skin cancers identified during the screening and age (r = 0.51, P < 0.001) and BMI (r = 0.373, P < 0.001).

The majority of surfers self-rated their surfing skill level as intermediate (59.2%) followed by advanced (36.7%), with the remainder self-classified as beginners (4.2%). All swimmers reported being recreational, whereas the majority of surfers were recreational (89.2%) followed by competing with their local board riding club (5.8%). The majority of surfers reported primarily riding a short board (56.7%), followed by a long-board (20.8%). Swimmers were more experienced (+13.1% years); however, surfers had greater UVR exposure (+34.9%, P = 0.172) as estimated via activity reported for the previous 12 months (total hours/year). Most surfers (93.3%) surfed year-round, with 100% of participants surfing during the summer months. There were fewer swimmers (65.7%) who swam year-round; however, all swimmers also swam during the summer months.

Both groups reported completing some activity during peak UVR (surfers 70.2% vs. swimmer 39.2%, P = 0.553). However, both groups were similar in their estimated percentage of selected aquatic activity completed during peak UV (surfers 42.8% vs. swimmers 45.8%, P = 0.553) (Table 1).

Prevention and screening practices

There was no difference with regard to the number of prevention strategies reported between surfers and swimmers (P = 0.380, 3.2 vs. 3.1, respectively). However, a significantly higher percentage of surfers reported using either a chemical and/or physical prevention strategy as compared to swimmers (100% vs. 92.7%, respectively; P = 0.003). With regard to scalp protection, a higher percentage of swimmers utilized a swim cap as opposed to surfers wearing a surf hat (43.6% vs. 20.7%, respectively; P = 0.002). Significantly more surfers utilised a rashie as opposed to swimmers (80.2% vs. 49.1%, respectively; P < 0.001); however, there were no differences between groups with regards to sunscreen, zinc, and lip balm use, or the reapplication of sunscreen (surfers 44.0% vs. swimmers 47.9%).

The majority of each group (surfers 63.8% vs. swimmers 72.2%) placed emphasis upon whom conducted the skin examinations, with most (surfers 43.9% vs. swimmers 54.2%) choosing to see a skin cancer doctor, followed by their GP (Table 2). The majority of participants (surfers 97.4% vs. swimmers 100%) were self-referred for their skin cancer screening in this study, with less than one-half of the participants in each group (surfers, 43.3%; swimmers 40.7%) having undergone a skin cancer check within the previous 12 months. There was also no difference identified between groups with regard to completing regular (self and/or partner) exams for suspicious lesions (surfers 36.2% vs. swimmers 35.2%; P = 0.897).

History and risk related to skin cancer

Both groups identified as having a history of skin cancer (surfers 41.4% vs. swimmers 36.4%, respectively) and a family history of skin cancer (surfers 52.6% vs. swimmers 43.6%). The majority of surfers and swimmers (74.1% vs. 69.0%) that experienced blistering sunburns as a child were significantly (P < 0.001) more likely found to have at least one skin cancer (BCC, SCC, SCC in situ, MSC) identified in this study than those who did not experience blistering sunburns as children. Surfers reported a significantly higher (P = 0.025) number of lesions of concern (37.1%) as compared to swimmers (20.0%) prior to being screened (Table 3).

There was no difference (P = 0.381) in years’ experience (years surfing or years swimming) or in the number of hours of aquatic activity per week (P = 0.075) between groups. With regard to weeks of aquatic activity reported per year, surfers were significantly (+10.5%, P = 0.026) more active than swimmers. Surfers over the year spent significantly (+35.3%, P = 0.038) more time (total hours per year) in their aquatic activity than swimmers. When lifetime aquatic activity was estimated (number of years’ experience × hours per year), there was no difference (P = 0.117) between groups, although surfers, with fewer years’ experience, completed more lifetime aquatic hours than swimmers (+38.9%). However, when groups were combined and aquatic activity expressed in quartiles, there were more participants identified with skin cancers when investigating total aquatic activity exposure (Fig. 1A, quartile 1 vs. quartile 4). Likewise, with groups combined, there were more skin cancers identified with increased activity exposure (Fig. 1B, quartile 1 vs. quartile 4).

Skin cancer screening results

The skin specialist identified a total of 74 participants with precancerous AKs during the screening (surfers 59, swimmers 15) and 110 skin cancers (BCC, SCC, SCC in situ, MSC). A significantly (P < 0.005) greater number of surfers were identified with a skin lesion (PSC, NMSC, MSC) during the screening than swimmers (132 vs. 52, respectively; OR 1.85, 95% CI [1.0–3.5]). Surfers also had a higher number (P = 0.076) of AK and rate (per 100,000) compared to swimmers (37,068 vs. 21,818 respectively, OR 1.7). Surfers also had a higher number of BCCs compared to swimmers (32 vs. 10, respectively); however, a lower rate (11,206 vs. 14,545 respectively, OR 0.77). Surfers had a slightly higher number of SCC in situ than swimmers (24 vs. 22; 13,793 vs. 12,727, OR 1.08). Surfers had a significantly (P = 0.05) higher number of SCCs however, a lower rate of SCCs compared to swimmers (1,724 vs. 3,636, OR 3.51). With regard to malignant melanomas, surfers had a higher rate (P = 0.137) as compared to the swimmers (5,172 vs. 1,818, OR 3.24).

When compared to the general Australian population, surfers and swimmers had higher ORs, which included BCCs (OR 7.27 vs. 9.44, respectively), SCCs (OR 1.67 vs. 3.51, respectively) and MSC (OR 96.7 vs. 34.0, respectively) (Table 4).

A MANCOVA demonstrated statistically significant differences in skin cancer totals (total number of skin cancers, total AKs, BCCs, SCC, SCC in situ, melanoma) based upon the history of skin cancer (F = 3.98, Pillai’s Trace = 1.40, P < 0.001) with history of skin cancer as a fixed factor and covariates of Fitzpatrick skin type and age. Adjusting the mean number of skin cancers observed for covariates (age and Fitzpatrick skin type), totals were higher for those with a history of skin cancer (mean with skin cancer history 7.24 vs. 3.95 with no skin cancer history, P = 0.012). A similar trend was observed for AKs (6.03 vs. 3.69, not significant (NS)), total BCC (0.66 vs. 0.06, P = 0.001), total SCC in situ (0.50 vs. 0.101, P = 0.042), total melanoma (0.62 vs. 0.44, NS) and total SCC (0.04 vs. −0.01, P = 0.024, with the negative mean value due to adjustment to the mean by covariates).

A total of 110 samples were sent for histopathology, and all (100%) samples were confirmed positive as either NMSC (BCC, SCC, SCC in situ) or MSC.

Fitzpatrick skin type and skin cancer

With regard to Fitzpatrick skin type and the number of skin cancers (P = 0.013), the majority of skin cancers (groups combined) were seen in the Fitzpatrick skin type 4 (59.1%), this was followed by Type 3 (19.9%) and Type 5 (15.8%) (Table 5). We did not have any participants with a Fitzpatrick type 1 or type 6 skin type. Surfers and swimmers presented with predominantly skin type 4 (57.8% vs. 61.8%, respectively), and this was followed by skin type 3 (21.6%) and skin type 5 (13.4%) in surfers and skin type 5 (20.0%) and skin type 3 (16.4%) in swimmers (Table 5).

Gender differences

There was no significant difference between genders with regard to experience between genders (P = 0.253); however, males had significantly more aquatic activity (weeks per year) (P = 0.036, +10.6%) than female participants. Females spent significantly more hours (P = 0.004, +57.2%) in their activity in peak UVR than male participants. There was no difference regarding the use of any prevention strategies between genders (males 96.8%, females 98.7%). More males reported experiencing sunburns as a child (59.7% vs. 63.8%), and a higher percentage of males reported having a sunburn in the previous 12 months (75.5% vs. 62.3%). A significantly (P < 0.001) higher number of females reported using a tanning bed (31.2% vs. 8.5%). More males reported a family history of skin cancer (P = 0.084, 44.7% vs. 33.8%). Males also had a significantly higher number of skin cancers (all lesions) as opposed to female participants (P = 0.011) and a significantly (P = 0.042) higher number of NMSC and MCS (combined).

Strengths and Limitations

This was the first study in Australia assessing the point prevalence of PSC, NMSC and MSC in surfers and swimmers via total body skin checks conducted by a clinician. Further strengths include our methodology included screening conducted by a specialist as opposed to participant recall via a survey. Additionally, the confirmation of all (100%) histopathology samples by commercial laboratory analysis matched the clinician’s findings.

Our participants had a high degree of homogeneity as they were from a limited geographic locale; however, this restrictive inclusion criteria limits the ability to extrapolate our findings to other surfers and swimmers within Australia, or elsewhere in the world. However, intermittent exposure to UVR is well documented in the literature as a casual mechanism for the development of NMSC and MSC, and we believe our findings have wide, geographic relevance.

It has, however, been reported that Fitzpatrick’s skin type is a confounding variable for the development of NSMC and MSC (Ting et al., 2007). Although Fitzpatrick’s skin type was assessed by a clinician, we cannot determine the effect Fitzpatrick skin type had on the development of NMSC and MSC in our participants, rather recognize that skin type is recognized as contributory to the development of skin cancer. Additionally, we did not note ethnicity, which may account for the skew in the skin types of participants in our study. Additionally, a significant number of the participants were very tanned with significant sun damage, making it difficult to assess Fitzpatrick skin type.

A limitation of the present study was that the sample size was relatively small. However, when compared to the very limited number of skin cancer screening studies that incorporated clinician screening (Dozier et al., 1997; Douglas, Baillie & Soyer, 2011), we did exceed the participant numbers in those previously published similar studies. We also recognize we did not account for confounding factors such as occupational and other UVR exposure (i.e., other outdoor activities), which would have contributed to the point prevalence’s reported in this study. As it was not possible to attain a list or number of surfers in the area, the participants were self-selected. Therefore, we were not able to utilize random sampling, and as such, selection bias may have occurred. As participants were self-referred to the study, it is possible we attracted participants who were more concerned about their health with regard to skin cancer, and this would represent an additional bias. Additionally, as the total number of surfers and swimmers in the region is unknown, we cannot assess nor calculate a response rate, and therefore determine the representativeness of our participants to their respective aquatic groups. We believe there is no confounding bias as we did not investigate casual relationships.