Participants were randomized into intervention and control groups.
eFigure. Skin Awareness Study Case Report Form
Janda M, Youl P, Neale R, Aitken J, Whiteman D, Gordon L, Baade P. Clinical Skin Examination Outcomes After a Video-Based Behavioral InterventionAnalysis From a Randomized Clinical Trial. JAMA Dermatol. 2014;150(4):372-379. doi:10.1001/jamadermatol.2013.9313
Older men are at risk of dying of melanoma.
To assess attendance at and clinical outcomes of clinical skin examinations (CSEs) in older men exposed to a video-based behavioral intervention.
Design, Setting, and Participants
This was a behavioral randomized clinical trial of a video-based intervention in men aged at least 50 years. Between June 1 and August 31, 2008, men were recruited, completed baseline telephone interviews, and were than randomized to receive either a video-based intervention (n = 469) or brochures only (n = 461; overall response rate, 37.1%) and were again interviewed 7 months later (n = 870; 93.5% retention).
Video on skin self-examination and skin awareness and written informational materials. The control group received written materials only.
Main Outcomes and Measures
Participants who reported a CSE were asked for the type of CSE (skin spot, partial body, or whole body), who initiated it, whether the physician noted any suspicious lesions, and, if so, how lesions were managed. Physicians completed a case report form that included the type of CSE, who initiated it, the number of suspicious lesions detected, how lesions were managed (excision, nonsurgical treatment, monitoring, or referral), and pathology reports after lesion excision or biopsy.
Overall, 540 of 870 men (62.1%) self-reported a CSE since receiving intervention materials, and 321 of 540 (59.4%) consented for their physician to provide medical information (received for 266 of 321 [82.9%]). Attendance of any CSE was similar between groups (intervention group, 246 of 436 [56.4%]; control group, 229 of 434 [52.8%]), but men in the intervention group were more likely to self-report a whole-body CSE (154 of 436 [35.3%] vs 118 of 434 [27.2%] for control group; P = .01). Two melanomas, 29 squamous cell carcinomas, and 38 basal cell carcinomas were diagnosed, with a higher proportion of malignant lesions in the intervention group (60.0% vs 40.0% for controls; P = .03). Baseline attitudes, behaviors, and skin cancer history were associated with higher odds of CSE and skin cancer diagnosis.
Conclusions and Relevance
A video-based intervention may increase whole-body CSE and skin cancer diagnosis in older men.
anzctr.org.au Identifier: ACTRN12608000384358
Melanoma is a common malignancy of the skin. In Australia in 2008, the age-standardized rate was 65 melanomas per 100 000 population,1,2 compared with 21 per 100 0003,4 in the United States. Although a stabilization of incidence rates in younger birth cohorts has been observed,5,6 the incidence in older age groups continues to increase in the United States, Australia, and Europe.7,8 In the United States, death rates from melanoma have decreased in women but have increased in men.9
Removing melanomas when they are thin (<1 mm) is associated with lower morbidity and mortality rates.10- 12 Early detection is an important strategy to reduce the burden of melanoma13 and can be achieved through visual inspection by a layperson (skin self-examination [SSE]) or a clinician (clinical skin examination [CSE]). In 1996, a population-based case-control study suggested that SSE was associated with a survival benefit.14 A case-control study in Queensland showed that melanomas detected during deliberate examinations (by a layperson or a physician) were thinner than those detected otherwise.15 Having 1 whole-body CSE within the past 3 years can reduce by 14% the risk of diagnosis of a thick melanoma.16 This may improve 10-year survival rates among screened (92.6%) vs unscreened (90.4%) melanoma survivors,16 although lead time bias needs to be considered. Clinical skin examinations have also been shown in other studies to demonstrate thin melanomas and reduce the incidence of thick melanomas.15,17- 20 A skin cancer screening project in Germany reported a reduction in mortality rates from melanoma in a state offering screening by CSE, compared with states not offering CSE screening.21
Approximately 30% of persons attend a physician for a CSE at least every 3 years,22 but older men are less likely than other populations to do so.16,21 In addition, other investigators have found that men have worse survival rates than women even after controlling for tumor thickness, suggesting that sex-specific biological factors may play a role in survival. Older men are also more likely to have diagnoses of thick melanomas, and their melanomas are more likely to be fatal.23 A cost analysis estimated that providing CSEs to men aged at least 50 years would incur health care costs similar to those of other early detection programs, such as mammography for breast cancer or fecal occult blood testing for colorectal cancer.24 Even so, melanoma screening is currently not recommended in most countries25 owing to lack of evidence of a mortality benefit in randomized trials (although 1 trial is currently ongoing).21,22
The present study forms part of a randomized behavioral trial of a video-based intervention designed to improve SSE, skin awareness, and CSE behaviors in men aged at least 50 years. Previous reports from this trial have focused on methods and SSE outcomes.26,27 This analysis focused on the prespecified secondary aim of the trial to assess CSE attendance and outcomes. We aimed to assess whether the intervention increased the proportion of men who presented to a physician for a CSE, received a whole-body CSE, and received a diagnosis of skin cancer. Another aim was to determine factors other than the intervention or control condition associated with having a CSE or skin cancer diagnosis during the trial.
Ethical approval was received from the Queensland University of Technology’s ethics committee (approval QUT 0600000645). Between June 1 and August 31, 2008, a total of 930 men aged at least 50 years were recruited through random selection from the Queensland electoral roll (response rate, 37.1%) (Figure). Eligibility criteria included proficiency in English, access to a digital video disc (DVD) player, and no previous history of melanoma. Participants were enrolled into a randomized clinical trial, the Skin Awareness Study (anzctr.org.au Identifier: ACTRN12608000384358). All participants provided written informed consent.
Intervention participants received both video-based and written skin awareness educational materials, and control group participants received only the written educational materials.27 The intervention was underpinned by the Health Belief Model.28 The video highlighted the seriousness of a melanoma diagnosis (perceived seriousness according to the health belief model), risk factors for melanoma, and the increased risk in men aged at least 50 years (perceived susceptibility); modeled a whole-body SSE (self-efficacy); presented a melanoma surgeon who encouraged SSE (cues to action) and presentation to a physician for a whole-body CSE; and showed a CSE being performed (overcoming barriers). A national sports personality along with melanoma survivors encouraged men to become skin aware (benefit).
Outcomes were the prevalence and frequency of having undergone any type or whole-body CSEs since baseline, as well as clinical and histopathological outcomes of skin lesions treated during the past 6 months. Overall, 469 men were randomized to the intervention group and 461 to the control group. Baseline telephone survey results were available for 929 participants.26,27 At baseline, 80.8% of men reported that a physician had ever checked any part of their skin for early signs of skin cancer, and 38.8% had undergone a whole-body CSE within the past 12 months.26
For the present analysis, we used data from a telephone interview administered 7 months after baseline, along with information from participants’ physicians. Participants were asked whether they had undergone CSE within the past 6 months. The validity of CSE self-report had been previously established (93.7% concordance between self-report and physician report for CSE within the past 3 years), with some evidence for telescoping when a shorter interval was assessed (74.3% concordance for CSE within the past 12 months).29 If participants reported having undergone a CSE, we asked about the type of CSE (skin spot, partial body, or whole body), who initiated it (the participant himself or his physician during a consultation for another reason), whether the physician noted any suspicious lesions, and, if so, how they were managed. With participant consent, we asked the physician to complete a case report form (eFigure in the Supplement) that included type of CSE, who initiated it, number of suspicious lesions detected, and how lesions were managed (nonsurgical treatment, surgical treatment [excision or biopsy], monitoring, or referral), and we obtained pathology reports for excisions or biopsies. Analysis was restricted to CSEs completed after the study starting date, October 1, 2008, and before the 7-month interview.
Analyses were performed using SAS software (versions 9.2 and 9.3; SAS Institute). Descriptive analyses were conducted, and χ2 tests and Wilcoxon rank sum tests were used to assess differences in self-reported outcomes between intervention and control groups; χ2 tests were also used to compare the distribution of physicians’ responses to each question in the case-report forms and diagnostic outcomes between treatment arms. Agreement between participant-reported and physician reported data was assessed using the Cohen κ statistic.
Bivariate logistic regression analyses were initially conducted, including demographic and clinical factors, phenotypic characteristics, SSE behaviors, and attitudes and social supports associated with undergoing at least 1 partial- or whole-body self-reported CSE during the study period. Multivariable logistic regression was then used to assess which characteristics were independently associated with self-reported CSE after adjustment for other variables (key demographic and skin cancer risk factors, sun protection behaviors, attitudes, and beliefs, as described elsewhere26), including randomization to intervention or control groups. Factors with a P value <.20 were initially included in the multivariable logistic regression, removed individually, and then reentered while we observed changes in the likelihood ratio to derive the most parsimonious model. Terms were retained if the P value was less than .05 within the multivariable model. Similarly, we established baseline factors independently associated with the diagnosis of skin cancer (melanoma, squamous cell carcinoma [SCC], or basal cell carcinoma [BCC]).
Once baseline interviews were complete (Table 1 presents baseline characteristics), participants were randomized into intervention and control groups stratified by area of residence (southeast corner of vs rest of Queensland); randomization was based on a computer-generated random number list generated separately from other study procedures by the study statistician (P. B.). Given the nature of the intervention, it was not possible to mask participants for their group assignment; however, telephone interviewers were working for a professional telephone survey company independent from the research team and masked to participants’ allocations. The 7-month follow-up telephone interviews were completed by 870 of 930 men (93.5% of those enrolled); the Figure summarizes participant flow. Demographic characteristics at baseline have been described elsewhere.26
Overall, at the 7-month interview, 475 of 870 men (54.6%) self-reported that a physician had deliberately checked any part of their skin during the past 6 months, and these results did not differ between intervention (246 of 436 [56.4%]) and control (229 of 434 [52.8%]) groups (P = .28). There was also no difference in the number of participants who reported that the physician looked at a skin spot during a consultation for another reason (intervention, 114 of 436 [26.1%]; control, 112 of 434 [25.8%]). However, participants in the intervention group (154 of 436 [35.3%]) were significantly more likely than controls (118 of 434 [27.2%]; P = .01) to report a whole-body CSE during the past 6 months. Among participants who reported either a dedicated CSE or skin spot check during another consultation, Table 2 compares the distribution of participants’ self-reported outcomes for these consultations. Men in the intervention group were more likely to have been asked by their physician to return for a follow-up examination (P = .001), but there was no difference between intervention or control groups in relation to self-reported skin lesion treatment (Table 2).
In the multivariable model, baseline factors positively associated with a self-reported CSE within the first 6 months of the trial included having a regular general practitioner (odds ratio, 1.49; 95% CI, 1.15-1.92), having had a spot or mole removed in the past (1.45; 1.24-1.71), current concern about a spot or mole (1.31; 1.10-1.56), having checked one’s own skin in the past 6 months (1.15; 1.00-1.33), having undergone CSE in the previous 12 months (1.47; 1.26-1.70), and sometimes or usually wearing a hat (1.34; 1.01-1.78). Within men in the intervention group who reported at least 1 CSE, those who watched the DVD more than once were more likely to report a whole-body CSE (62.2%) than those who watched the DVD once (55.2%) or did not watch it (50.0%); however, this difference was not statistically significant (P = .34).
Of men who reported undergoing CSE in the previous 6 months, 321 of 540 (59.4%; 159 in the intervention and 162 in the control group) gave consent for their physician to be contacted by the study team for further details about the CSE. Men who had black hair, no previous history of skin excision or treatment, and no CSE within the 12 months before baseline were less likely to consent for their physician to be contacted (all P < .05). Men who did not provide consent to contact their physician were less likely to self-report that at least 1 skin lesion was found during the CSE (84 of 216 [38.9%]) than those who gave permission (165 of 321 [51.4%]; P = .004), and they self-reported a lower distribution of lesions requiring treatment (median, 2 lesions; range, 1-15) than men who consented to physician contact (median, 2 lesions; range, 1-28; P < .001).
In total, medical case report forms and pathology reports (where applicable) were obtained from the physician for 266 of the 321 men (82.8%) who consented. Of these case report forms, 211 of 266 (79.3%) were for CSEs conducted within the study period and were used in this analysis (104 in the intervention and 107 in the control group).
Based on the case reports received from physicians, men in the intervention group were more likely to have undergone a whole-body CSE than those in the control group (74.5% vs 61.4%; P = .046); however, men in both groups were equally likely to be perceived by the physician as having initiated the CSE (64.7% vs 57.9%; P = .31). After the CSE, physicians treated, monitored, or referred at least 1 lesion in 76% of participants (76.0% in the intervention and 76.6% in the control group). Of those, 49.3% of participants (104 of 211) had nonsurgical management of at least 1 lesion (50 of 104 [48.1%] in the intervention and 54 of 107 [50.5%] in the control group). Many of them (86 of 211 [40.8%]) were treated with cryotherapy. Overall, 34.1% (72 of 211) underwent surgical excision or biopsy of at least 1 lesion (41.3% in the intervention and 27.1% in the control group; P = .03), with a median of 2 lesions found (Table 3). The concordance between self-reported and physician-reported CSE was moderate for whole-body CSE (Cohen κ = 0.53) and for management of any lesions (Cohen κ = 0.43).
Pathology reports were obtained for 130 lesions that were excised or sampled for biopsy (85 in the intervention and 45 in the control group). Overall, 2 melanomas, 29 SCCs, 38 BCCs, 17 solar keratoses, 3 dysplastic nevi, 9 benign nevi, and 32 other pigmented or nonpigmented lesions were diagnosed. The 2 melanomas were diagnosed in intervention participants. Thus, the study obtained a melanoma detection rate of 2 of 469 (426 per 100 000). In addition, 21 SCCs and 28 BCCs were detected in 104 intervention participants, and 8 SCCs and 10 BCCs in 107 control participants. Significantly more skin cancers were detected in the intervention group than the control group (60.0% vs 40.0%, respectively; P = .03) (Table 4).
Factors positively associated with a skin cancer diagnosis during the trial included being an intervention participant (odds ratio, 1.45; 95% CI, 1.20-2.08), conducting SSE within the past 6 months (1.60; 1.04-2.48), history of treatment for a spot or mole (1.78; 1.19-2.67), and self-reported CSE within the past 12 months (2.52; 1.21-5.23). Men who rarely or never stayed in the shade and men who tanned without burning were more likely to have a skin cancer diagnosed (odds ratio, 1.63 [95% CI, 1.10-2.43] and 3.24 [1.42-7.38], respectively) (Table 5).
Although screening for melanoma by CSE for men aged at least 50 years may be cost-effective,30 it is often not recommended owing to the absence of evidence of a mortality benefit in randomized clinical trials. However, data are accumulating from observational studies on the value of CSE for reducing melanoma thickness at diagnosis and mortality rates, highlighting the benefit for men aged at least 50 years.16,31 This study found that a video-based intervention designed to increase skin awareness, SSE, and presentation to a physician with suspicious skin lesions among men aged at least 50 years resulted in a higher prevalence of self- and physician-reported whole-body CSE than the provision of written materials alone among men who underwent any type of CSE. Among men who underwent CSE, 34.1% had excision or biopsy of at least 1 lesion, consistent with high levels of clinical suspicion for these lesions and highlighting the potential value of facilitating CSEs in this group of older men in Australia.
Compared with the control group, men receiving the video intervention were more likely to self-report undergoing whole-body CSE. Also noted by the physicians, a larger proportion of CSEs in intervention participants (74.5%) were whole-body examinations, which were recommended in the video intervention to make certain that lesions on difficult-to-see body areas were also assessed.6,32,33 Our analysis shows that men were more likely to self-report CSEs if they had a regular physician, previous SSEs and/or CSEs, previous treatment of skin lesions or moles, or current concern about a skin lesion, largely similar to previous findings.34,35 The complementary nature of SSE and CSE has been noted elsewhere in an investigation of skin cancer early detection behavior among melanoma survivors.36
A previous trial of a video-based intervention (the Check-It-Out trial),37,38 compared SSE and CSE outcomes among 1356 men and women (median age, 52 years). The intervention included educational materials provided in paper-based and video formats plus individual behavioral counseling (1 face-to-face and 1 telephone session). Control participants received the same attention but were counseled about healthy diet. Participants randomized to the SSE group were significantly more likely to undergo skin surgery during the first 6 months after the intervention (8% vs to 4% in the diet group). The number of malignant lesions found was small compared with our study (1 melanoma, 10 BCCs, and 3 SCCs), probably because of the younger age group involved and the lower skin cancer risk in the United States compared with Australia.8,38
Another trial that focused on improving early detection of skin cancers in men aged at least 50 years randomized men to receive or not receive photographs of their skin to help detect any changes in lesions.39 During the 2-year follow-up period, 34% underwent skin excision, similar to the 34.1% rate observed in our study. The proportions of cancers among the excised lesions (58% in the intervention and 42% in the control group) were also similar to those observed in our study. Hanrahan et al39 discussed whether this between-group difference in overall excised lesions may have reflected missed lesions in the absence of photographs or treatment of lesions by cryotherapy in the control participants. However, in our study, the proportions of participants treated with cryotherapy were similar between the 2 groups.
Our findings indicate that in 76.3% of men with a CSE reported by their physician, skin lesions were discovered that required some form of management, of which 40.0% and 60.0% were identified as skin cancers in the pathology report in the control and intervention groups, respectively. This may suggest that a targeted educational program such as ours may lead to early detection of melanoma or other skin cancers. Although the overall level of excisions may seem high, we reported elsewhere that Australian general practitioners are excellent at diagnosing skin cancer, needing to excise a mean of just 2 skin lesions to find 1 skin cancer.40 Furthermore, Fransen et al41(p566) reported that “83% of NMSC [nonmelanoma skin cancer] treatments were administered in people aged 55 years and over, and nearly two-thirds of NMSC treatments were administered in persons aged 65 years and over.”
Strengths of our study include its focus on men aged at least 50 years, a group at increased risk of dying of melanoma. Its limitations include the fact that the men who agreed to participate in the trial were already relatively skin aware at baseline (39% self-reported undergoing whole-body CSE within the 12 months before enrollment, with no difference between intervention and control groups). Our results could therefore underestimate the true effect of our intervention program, if less health-aware men are assumed to be more likely to have unidentified skin cancers. A relatively low proportion of men gave consent for us to contact their physicians (321 of 540 [59.4%]), mostly out of reluctance to create work for their physician, although the response rate from physicians was good (266 of 321 [82.9%]). This meant that CSE outcomes were available for 266 of 540 participants (49.3%) self-reporting a CSE. It is therefore likely that additional cancers were diagnosed but not recorded during the study. Men with fair phenotypes and previous skin excisions were more likely to consent to our contacting their physician. Compared with men who consented to physician follow-up, men who did not consent self-reported fewer lesions being found during CSEs. If men who did not give consent were at lower risk of skin cancer, our results may overestimate somewhat the number of skin cancers that could be diagnosed. As noted elsewhere,28 Skin Awareness Study participants may have been more health conscious than men from the general population, and 81.7% reported at baseline having ever undergone any type of skin examination by a physician. Our results may therefore overestimate what could be achieved in less health-conscious men.
In summary, our trial showed that men aged at least 50 years responded favorably to video-based education, increasing their skin awareness and attendance at whole-body CSE during 7 months of follow-up. Among men in both intervention and control groups, many malignant lesions were diagnosed and treated because of CSEs. We acknowledge that routine use of CSE as a screening tool will place a burden on the health care system and could lead to the detection of skin cancers that are relatively indolent and may never cause death or significant morbidity.22,25 However, with increasing evidence from observational studies supporting the effects of CSE in reducing the incidence of thick melanomas and melanoma-associated mortality rates16,31,42 and with evidence of potential reductions in the cost-benefit ratio,24 our results support implementing behavioral interventions to encourage skin awareness among men aged at least 50 years.
Accepted for Publication: October 26, 2013.
Corresponding Author: Monika Janda, PhD, School of Public Health and Institute of Health and Biomedical Innovation, Queensland University of Technology, Kelvin Grove, Brisbane, Queensland 4059, Australia (firstname.lastname@example.org).
Published Online: February 19, 2014. doi:10.1001/jamadermatol.2013.9313.
Author Contributions: Drs Janda and Gordon had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Janda, Youl, Neale, Gordon, Baade.
Acquisition of data: Janda, Baade.
Analysis and interpretation of data: Janda, Youl, Aitken, Whiteman, Baade.
Drafting of the manuscript: Janda.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Janda, Baade.
Obtained funding: Janda, Youl.
Administrative, technical, or material support: Janda, Youl, Aitken, Baade.
Study supervision: Janda, Whiteman, Baade.
Conflict of Interest Disclosures: None reported.
Funding/Support: This trial was funded by the Australian National Health and Medical Research Committee (project grant 497200; career development fellowships 1045247 to Dr Janda, 552404 to Dr Neale, and 1005334 to Dr Baade; principal research fellowship to Dr Whiteman; and public health early career fellowship 496714 to Dr Gordon).
Role of the Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank Jannah Baker, MBChB, PGDipSci, PGDipPH, for assistance with data analysis and Caitlin Horsham, BPH, Linda Finch, MPsych(Clin), and Josephine Auster, MPH, for research assistance. These individuals are affiliated with the School of Public Health, Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia, and were compensated by project grant 497200.