Personalized melanoma risk assessments and tailored prevention advice: a pragmatic randomized controlled trial in Australian general practice

Personalized melanoma risk assessments and tailored prevention advice: a pragmatic randomized... Abstract Background Personalized risk assessments using prediction models that incorporate several melanoma risk factors may promote melanoma-prevention behaviours. Objectives To evaluate the effect on short-term melanoma-prevention behaviours of web-based, real-time, model-generated personalized melanoma risk information and tailored prevention advice, and its feasibility and clinician acceptability. Methods Between February and April 2016, in an open randomized controlled trial across four general medical practices in New South Wales, Australia, 272 patients were randomly allocated to receive (i) real-time model-generated personalized melanoma risk assessment and tailored melanoma-prevention advice or (ii) generic melanoma-prevention advice. We measured self-reported melanoma-prevention behaviours at baseline and 6 weeks and the intervention’s feasibility and acceptability. Results Follow-up questionnaires were completed by 185 patients at 6 weeks: 174 assessed as average risk and 11 as high or very high risk. There were no statistically significant differences between intervention and control patients in sun protection, sun exposure or early diagnosis behaviours. When stratified by melanoma risk, average risk patients in the intervention group appeared to show greater sun protection at 6 weeks (mean difference = 0.23, on a scale of 1–5; 95% confidence interval: 0.01 to 0.45; P = 0.04) than patients in the control group; the P value for interaction between intervention and risk category was 0.10. There was favourable feedback from patients and general practitioners. Conclusions Web-based delivery in general practice of real-time, model-generated personalized melanoma risk prediction and tailored melanoma-prevention advice is feasible and acceptable. An apparent increase in sun protection behaviour in average risk patients warrants further evaluation in different risk groups. Health promotion, melanoma, primary health care, primary prevention, risk reduction behaviour, secondary prevention Introduction Primary prevention interventions based on sun protection can reduce the melanoma burden (1,2). Despite community-wide mass media campaigns (3), sun protection levels remain relatively low (4–9). Risk prediction models, which provide a personalized estimate of risk based on a combination of melanoma risk factors, and tailoring prevention interventions to melanoma risk (10,11) may motivate greater adoption of sun protection behaviours. Few melanoma risk prediction models have undergone external validation in independent populations or prospective evaluation in routine clinical settings (12,13). Based on data from an Australian population–based case–control study, we have developed a melanoma risk prediction model of self-assessed melanoma risk factors: hair colour, naevus density, first-degree family history, previous non-melanoma skin cancer and lifetime sunbed use (14). External validation of our model showed good discrimination between those with and without melanoma (area under the receiver operating curve between 0.63 and 0.67 across four independent population-based studies), good calibration and higher net benefit than obtained by simply classifying everyone as high or low risk (14). We aimed to evaluate prospectively the effect on short-term melanoma-prevention behaviours of delivering real-time model-generated personalized melanoma risk assessments and tailored prevention advice in general practice, and to assess the feasibility and ‘clinician acceptability’ of using the model in practice. Methods Design, setting and participants We conducted an open, individually randomized controlled trial with 1:1 allocation of participants to one of two parallel groups in four general practices (three urban and one rural, comprising 29 general practitioners) in New South Wales, Australia, from 16 February to 30 April 2016. Consecutive patients were invited from the general practice waiting rooms on different days and at different times to participate. Eligible patients were aged over 18 years and had the capacity to give informed consent in English. We excluded patients with a previous diagnosis of cutaneous melanoma. Intervention Before they saw the general practitioner, all eligible consenting patients completed a baseline electronic questionnaire delivered using a tablet computer and a web-based application developed for this study (Supplementary Figure S1). The web-based application randomly allocated patients to the intervention or control group, and delivered to intervention group patients a real-time, on-screen and printed model-generated personalized melanoma risk assessment and tailored prevention advice. Patients in each group were given a booklet containing generic melanoma-prevention information on risk factors and prevention measures. The melanoma risk assessments and tailored prevention advice were not given to the doctors separately from the patients. The risk assessments were communicated as absolute remaining lifetime risk of melanoma (to 85 years old) with a 100-person pictograph, a relative remaining lifetime risk (relative to average remaining lifetime risk in people of the same 5-year age and sex stratum in New South Wales), and a risk category (average, high or very high) (Fig. 1). Since most Australians have fair skin and almost all epidemiological studies evaluating melanoma risk factors have been conducted in fair-skinned populations, patients at low melanoma risk were allocated to the average risk category. We defined the risk category cut-points and tailored the prevention advice to existing cut-points specified in the Australian guidelines for preventive activities in general practice, which are based on published relative risks for individual risk factors (15). This advice recommended sun protection for those at average risk (relative risk ≤2), sun protection and a full-body clinical examination for those at high risk (relative risk between 2 and 6), and sun protection, a full-body clinical examination and full-body skin self-examination for those at very high risk (relative risk ≥6). Figure 1. View largeDownload slide View largeDownload slide View largeDownload slide Model-generated melanoma risk assessments and tailored melanoma–prevention advice from February to April 2016 Figure 1. View largeDownload slide View largeDownload slide View largeDownload slide Model-generated melanoma risk assessments and tailored melanoma–prevention advice from February to April 2016 Outcomes The primary outcome was short-term sun protection behaviours—seeking shade, sunscreen use, wearing sunglasses, wearing a hat and protective clothing—which were individually measured and summarized in a validated composite measure (16). Secondary outcomes included sun exposure, full-body clinical- and skin self-examinations, behavioural intentions, melanoma risk perception, feasibility and clinician acceptability. Data collection Patients were contacted by their preferred method (email, post or telephone) 6 weeks after they had completed the baseline questionnaire. Non-responders were sent up to three reminders at 2-week intervals. The baseline questionnaire contained questions on sociodemographic characteristics, self-assessed melanoma risk factors in our melanoma risk prediction model (14) and validated questions on sun protection behaviours, sun exposure, clinical- and skin self-examinations, melanoma-prevention intentions and melanoma risk perception (16–18). The follow-up questionnaire contained the same questions (Supplementary Figure S1). Sun protection behaviours were measured on a five-point Likert scale with reference to hours between 9 am and 5 pm during the last week, ranging from 1 corresponding with ‘never’ to 5 corresponding with ‘always’ and were averaged to form a composite measure. Patients were also asked to indicate their usual time spent outdoors between 9 am and 5 pm during the last week, and whether they had a clinical examination and a skin self-examination during the last 12 months (baseline) or the last 6 weeks (follow-up). Melanoma-prevention intentions, based on the trans-theoretical model of behaviour change (19), were measured on a five-point Likert scale with reference to hours between 9 am and 5 pm during the last week, ranging from 1 corresponding with ‘I have never thought of doing this’ (pre-contemplation) to 5 corresponding with ‘I have been doing this for quite a while’ (maintenance) and were averaged to form a composite measure. Melanoma risk perception was also measured on a five-point Likert scale using the question ‘How would you rate your risk of developing melanoma compared to other men or women of your age?’; ranging from 1 ‘much lower’ to 5 ‘much higher’ than average risk. We measured the time taken to complete the baseline electronic questionnaire, and time taken for intervention group patients to review their personalized melanoma risk assessments and tailored prevention advice on the web-based application. Intervention group patients were asked to evaluate the web-based risk assessments (Supplementary Material 2). Participating general practitioners were asked to evaluate the intervention after the study by responding to a questionnaire (Supplementary Material 3). Sample size Our power calculations indicated that we needed to randomize 262 participants to detect a minimum difference between the control and intervention group of 0.2 in the composite sun protection behaviours scale (range: 1–5), assuming a standard deviation (SD) of 2.5, a 5% two-sided significance level, 80% power and 25% lost to follow-up (20–23). The sample sizes were determined using PS software (24). Randomization Patients were randomized using a computer-generated schedule with randomly varied block sizes within each 5-year age and sex stratum. Allocation was assigned by the web-based application and concealed. Patients were not informed of their group; however, blinding of the researchers was not possible. Statistical methods Descriptive statistics were calculated for all variables. Sun protection behaviour, sun exposure, melanoma-prevention intentions and melanoma risk perception scores were analysed as continuous variables using analysis of covariance adjusted for age, sex, recruitment site and corresponding baseline value. Clinical- and skin self-examinations (dichotomized to adherent versus non-adherent to guidelines for preventive activities (15) based on estimated personalized melanoma risk) were analysed using log-binomial models adjusted for age, sex, site of recruitment and corresponding baseline value. We compared the baseline characteristics between those who were lost to follow-up and those who completed the follow-up questionnaire. Data were analysed on an intention-to-treat basis using SAS 9.4 statistical software with a two-sided significance level of 0.05. We report methods and results in accordance with the Consolidated Standards of Reporting Trials statement (25). Results Participants We approached 411 general practice patients; 355 (86.3%) who agreed to participate were assessed for eligibility. Among the 320 (90.1%) patients who were eligible and gave consent, 48 (15.0%) were called in to see their doctor before completing the baseline electronic questionnaire; therefore, 272 patients were randomized (Fig. 2). The baseline sociodemographic characteristics and melanoma risk profiles were similar in the two groups (Table 1). Among the 134 intervention group patients, the mean (SD) absolute remaining lifetime melanoma risk was 3.11% (5.0%) and mean relative remaining lifetime risk was 0.90 (1.15). Among the 138 control group patients, the mean (SD) absolute remaining lifetime melanoma risk was 2.94% (4.73%) and the mean (SD) relative remaining lifetime risk was 0.80 (1.05). Table 1. Baseline characteristics for general practice patients who were offered generic and personalized melanoma risk information from February to April 2016 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 aPictogram shown in Supplementary Figure 1 (question 9). View Large Table 1. Baseline characteristics for general practice patients who were offered generic and personalized melanoma risk information from February to April 2016 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 aPictogram shown in Supplementary Figure 1 (question 9). View Large Figure 2. View largeDownload slide Flow of participants from February to April 2016 through each stage of the randomized controlled trial Figure 2. View largeDownload slide Flow of participants from February to April 2016 through each stage of the randomized controlled trial Outcomes Melanoma-prevention behaviours, intentions and risk perception Six-week follow-up questionnaires were completed by 185 (68.0%) patients, 174 (94.1%) assessed as average melanoma risk and 11 (5.9%) as high or very high melanoma risk. There were no overall significant differences at follow-up between intervention and control patients in sun protection (P = 0.13), sun exposure (P = 0.82 for weekday exposure and P = 0.33 for weekend exposure), early diagnosis behaviours (P = 0.98 for clinical examination and P = 0.09 for skin self-examination), melanoma-prevention intentions or melanoma risk perception (Table 2). There were no new full-body clinical examinations and six new full-body skin self-examinations (five intervention group patients assessed as average melanoma risk and one control group patient assessed as very high melanoma risk). When stratified by risk category, sun protection behaviours (composite score) in those at average risk of melanoma were higher in intervention than control patients [mean difference = 0.23, on a scale of 1–5; 95% confidence interval (CI): 0.01 to 0.45; P = 0.04], as was use of sunglasses (mean difference = 0.43, on a scale of 1–5; 95% CI: 0 to 0.86; P = 0.05). The P values for the interactions of intervention by risk group were P = 0.10 for sun protection and P = 0.50 for sunglasses; however, the study was underpowered to detect meaningful interactions between risk groups and sun protection behaviours. Table 2. Differences between general practice patients who were offered personalized or generic melanoma risk information at 6-week follow-up from February to April 2016, adjusting for age, sex, site of recruitment and corresponding baseline value Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA NA, not applicable; SD, standard deviation. aRange of values was 1–5 (1 = never and 5 = always). bThe Likert scale responses were averaged to form a composite scale. cRange of values was 1–5 (1 = I have never thought of doing this and 5 = I have been doing this for quite a while). dRange of values was 1–5 (1 = much lower than average and 5 = much higher than average). eRange of values was 1–5 (1 = not at all concerned and 5 = very concerned). fAt baseline, patients were asked whether they had a skin examination in the last 12 months. At follow-up, the question referred to the last 6 weeks only. View Large Table 2. Differences between general practice patients who were offered personalized or generic melanoma risk information at 6-week follow-up from February to April 2016, adjusting for age, sex, site of recruitment and corresponding baseline value Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA NA, not applicable; SD, standard deviation. aRange of values was 1–5 (1 = never and 5 = always). bThe Likert scale responses were averaged to form a composite scale. cRange of values was 1–5 (1 = I have never thought of doing this and 5 = I have been doing this for quite a while). dRange of values was 1–5 (1 = much lower than average and 5 = much higher than average). eRange of values was 1–5 (1 = not at all concerned and 5 = very concerned). fAt baseline, patients were asked whether they had a skin examination in the last 12 months. At follow-up, the question referred to the last 6 weeks only. View Large Feasibility Patients took on average 5 minutes and 14 seconds to complete the baseline electronic questionnaire. Intervention group patients took on average 1 minute and 35 seconds to review their personalized melanoma risk assessments and 47 seconds to review their tailored prevention advice on the web-based application. Intervention patients reported the web-based melanoma risk assessment as easy to use (96%), easy to understand (97%) and useful (90%). Clinician acceptability We mailed a short questionnaire to 29 general practitioners who worked in the participating general practices; six (21%, two men and four women, aged 31–62 years old) completed it. All six general practitioners reported that the melanoma risk assessments and tailored prevention advice generated by the web-based application were easy to understand and unlikely to make the patients feel upset or worried. Five (83%) said they would be likely to use the web-based application if it became widely available. Conclusions The delivery of real-time model-generated personalized melanoma risk and tailored prevention advice in general practice using a web-based melanoma risk prediction model is highly feasible and acceptable. Although we did not find any overall differences between all intervention and control patients in sun protection, sun exposure, early diagnosis behaviours, melanoma-prevention intentions or risk perception at 6-week follow-up, there were modest increases in sun protection behaviours in intervention patients compared with control patients among those at average melanoma risk. We could find only three previously reported instances of use of melanoma risk prediction models in general practice (26–28). One of the studies (Rat et al. 2014) (28), conducted in western France, appeared similar to ours. It evaluated the effect of model-generated melanoma risk assessments (29) on sun protection behaviours in a clustered-randomized controlled trial where general practitioners were randomized to deliver targeted screening and education intervention or conventional information-based campaign to 173 patients identified as at high risk. At 5-month follow-up, high-risk patients in the intervention group were less likely to sunbathe and more likely to have performed skin self-examinations compared with the high-risk patients in the control group (28). It is not clear in this study whether patients in the intervention group were aware of their high risk, as they were in our study, and, if they were, whether this knowledge contributed to their more favourable outcomes. The risk category classifications and proportion of patients classified as high risk were quite different between the study of Rat et al., which classified high risk using a model relative risk category cut-point of 11 and 46% of patients as at high risk, and our study, which used relative risk category cut-points from the Australian guidelines for preventive activities in general practice (15) and classified 6% of patients as high or very high risk. It is worth noting that the risk category cut-points given in the Australian guidelines were based on individual risk factors, not on a combined risk factor model. If we instead classified our participants based on their individual risk factors using the same relative risk cut-points (2 for high and 6 for very high risk), we would have classified 19% of patients as at high or very high risk. Choosing appropriate cut-points to define levels of risk is not straightforward and depends on an assessment of the net benefit of different cut-points and the relative benefits and harms of the intervention (14). These issues need further investigation before they can be readily used to inform the choice of cut-points for classifying high risk. While the approach we used was objective, it was a disadvantage that the small numbers of patients classified as high or very high risk substantially diminished its power to evaluate interaction by subgroup. To our knowledge, ours is the first study to deliver real-time model-generated personalized melanoma risk and tailored prevention advice to patients across all risk profiles. Its main strengths are the randomized controlled trial design and the pragmatic evaluation in routine general practice, in which 86% of approached patients agreed to participate. The study also included practices from both urban and rural areas to represent the breadth of the population. Furthermore, we used a novel method of data collection and risk profile delivery, using a web-based application on a tablet computer to minimize time taken and transcription errors. Randomization was also facilitated within the application using a custom-built computer algorithm. It was a potential limitation that we recruited more women than men; however, this reflects the general practice consulting population (30). While the general practitioner participation in evaluating the intervention was low, it was in line with other studies (31,32). Thirty-two per cent of patients in the randomized controlled trial were lost to follow-up, which might have introduced selection bias. We were not able to collect information from non-responders, but those who were lost to follow-up had similar baseline characteristics to those who completed the follow-up questionnaire (Supplementary Material 4). We recruited participants from summer to early autumn and completed the follow-up in winter when solar ultraviolet radiation is lower; however, while season and weather influence sun protection behaviours, changes in season would have affected the control and intervention groups equally. We delivered the intervention directly to patients and did not measure whether the patients discussed the melanoma risk information and prevention advice with their general practitioners, which would require an audit of the medical records. We also did not measure melanoma incidence or mortality as outcomes; which would require a much larger study with very long follow-up. Tailored prevention advice based on real-time model-generated risk assessment in general practice is highly feasible and acceptable and suggests that similar approaches to offer prevention, screening or support that is tailored to the characteristics and needs of the patient could be used more widely for other diseases in primary and secondary care settings in Australia and other countries. Low melanoma-prevention behaviours were observed in a recent international cross-sectional survey conducted across 23 countries, with higher prevention behaviours among people from countries with higher ambient ultraviolet radiation like Australia, Chile and Greece (33). Our findings provide some evidence for increased short-term sun protection behaviours among those at average melanoma risk who received melanoma risk prediction and tailored prevention advice. Future studies should be large enough to analyse within each risk group and probably use a lower absolute risk category cut-point to define greater numbers of patients as at high risk. Meanwhile, more research is needed to establish a broadly applicable, objective basis for defining high risk when using risk prediction models in preventive interventions. Supplementary Material Supplementary data are available at Family Practice online. Declaration Funding: KV was supported by a University of Sydney Postgraduate Scholarship in Cancer Epidemiology (funded through AEC’s Cancer Institute NSW fellowship), a Sydney Catalyst Top-Up Research Scholar Award, and a Primary Care Collaboration Cancer Clinical Trials Group and Clinical Oncology Society of Australia Training Award in Cancer and Primary Care. AEC was supported by fellowships from the Cancer Institute NSW (15/CDF/1–14) and the National Health and Medical Research Council (NHMRC) (1147843). The funding organizations had no role in the study design, data collection, analysis or interpretation of data, and writing or submitting the manuscript. Ethical approval: The study was approved by The University of Sydney Human Research Ethics Committee (2014/144) and prospectively registered with the Australian New Zealand Clinical Trials Registry (ACTRN12615001019594) available at https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?ACTRN=12615001019594. Conflict of interest: None declared. Acknowledgements We wish to thank patients and staff at Montrose Medical Practice, Gordon Family Medical Practice, Summer Hill Village Medical Practice and Sussex Inlet Medical Centre for their participation. We also wish to thank Allison Grech, Brooke Beswick and Amelia Smit for their help with recruitment and data management. References 1. Armstrong BK , Kricker A . How much melanoma is caused by sun exposure ? Melanoma Res 1993 ; 3 : 395 – 401 . Google Scholar CrossRef Search ADS PubMed 2. Olsen CM , Wilson LF , Green AC , et al. Cancers in Australia attributable to exposure to solar ultraviolet radiation and prevented by regular sunscreen use . Aust N Z J Public Health 2015 ; 39 : 471 – 6 . Google Scholar CrossRef Search ADS PubMed 3. Wu YP , Aspinwall LG , Conn BM , et al. A systematic review of interventions to improve adherence to melanoma preventive behaviors for individuals at elevated risk . Prev Med 2016 ; 88 : 153 – 67 . Google Scholar CrossRef Search ADS PubMed 4. Montague M , Borland R , Sinclair C . Slip! Slop! Slap! and SunSmart, 1980–2000: skin cancer control and 20 years of population-based campaigning . Health Educ Behav 2001 ; 28 ( 3 ): 290 – 305 . Google Scholar CrossRef Search ADS PubMed 5. Sinclair C , Foley P . Skin cancer prevention in Australia . Br J Dermatol 2009 ; 161 ( suppl 3 ): 116 – 23 . Google Scholar CrossRef Search ADS PubMed 6. Dobbinson S , Wakefield M , Hill D , et al. Prevalence and determinants of Australian adolescents’ and adults’ weekend sun protection and sunburn, summer 2003–2004 . J Am Acad Dermatol 2008; 59 ( 4 ): 602 – 14 . CrossRef Search ADS PubMed 7. Dobbinson SJ , Wakefield MA , Jamsen KM , et al. Weekend sun protection and sunburn in Australia trends (1987–2002) and association with SunSmart television advertising . Am J Prev Med 2008 ; 34 : 94 – 101 . Google Scholar CrossRef Search ADS PubMed 8. Volkov A , Dobbinson S , Wakefield M , Slevin T . Seven-year trends in sun protection and sunburn among Australian adolescents and adults . Aust N Z J Public Health 2013 ; 37 : 63 – 9 . Google Scholar CrossRef Search ADS PubMed 9. Koch S , Pettigrew S , Minto C , et al. Trends in sun-protection behaviour in Australian adults 2007–2012 . Australas J Dermatol 2017 ; 58(2): 111–6. 10. Ahmed H , Naik G , Willoughby H , Edwards AG . Communicating risk . BMJ 2012 ; 344 : e3996 . Google Scholar CrossRef Search ADS PubMed 11. Freedman AN , Seminara D , Gail MH , et al. Cancer risk prediction models: a workshop on development, evaluation, and application . J Natl Cancer Inst 2005 ; 97 : 715 – 23 . Google Scholar CrossRef Search ADS PubMed 12. Vuong K , McGeechan K , Armstrong BK , Cust AE . Risk prediction models for incident primary cutaneous melanoma: a systematic review . JAMA Dermatol 2014 ; 150 : 434 – 44 . Google Scholar CrossRef Search ADS PubMed 13. Wernli KJ , Henrikson NB , Morrison CC , et al. Screening for skin cancer in adults: updated evidence report and systematic review for the US preventive services task force . JAMA 2016 ; 316 : 436 – 47 . Google Scholar CrossRef Search ADS PubMed 14. Vuong K , Armstrong BK , Weiderpass E , et al. ; Australian Melanoma Family Study Investigators . Development and external validation of a melanoma risk prediction model based on self-assessed risk factors . JAMA Dermatol 2016 ; 152 : 889 – 96 . Google Scholar CrossRef Search ADS PubMed 15. RACGP . Guidelines for Preventive Activities in General Practice . 9th edn . East Melbourne, Victoria : The Royal Australian College of General Practitioners , 2016 . 16. Glanz K , Yaroch AL , Dancel M , et al. Measures of sun exposure and sun protection practices for behavioral and epidemiologic research . Arch Dermatol 2008 ; 144 : 217 – 22 . Google Scholar CrossRef Search ADS PubMed 17. O’Riordan DL , Nehl E , Gies P , et al. Validity of covering-up sun-protection habits: association of observations and self-report . J Am Acad Dermatol 2009 ; 60 : 739 – 44 . Google Scholar CrossRef Search ADS PubMed 18. Bränström R , Kristjansson S , Ullén H , Brandberg Y . Stability of questionnaire items measuring behaviours, attitudes and stages of change related to sun exposure . Melanoma Res 2002 ; 12 : 513 – 9 . Google Scholar CrossRef Search ADS PubMed 19. Prochaska JO , Velicer WF . The transtheoretical model of health behavior change . Am J Health Promot 1997 ; 12 : 38 – 48 . Google Scholar CrossRef Search ADS PubMed 20. Robinson JK . Behavior modification obtained by sun protection education coupled with removal of a skin cancer . Arch Dermatol 1990 ; 126 : 477 – 81 . Google Scholar CrossRef Search ADS PubMed 21. Glanz K , Schoenfeld ER , Steffen A . A randomized trial of tailored skin cancer prevention messages for adults: project SCAPE . Am J Public Health 2010 ; 100 : 735 – 41 . Google Scholar CrossRef Search ADS PubMed 22. Falk M , Magnusson H . Sun protection advice mediated by the general practitioner: an effective way to achieve long-term change of behaviour and attitudes related to sun exposure . Scand J Prim Health Care 2011 ; 29 ( 3 ): 135 – 43 . Google Scholar CrossRef Search ADS PubMed 23. Glazebrook C , Garrud P , Avery A , Coupland C , Williams H . Impact of a multimedia intervention “Skinsafe” on patients’ knowledge and protective behaviors . Prev Med 2006 ; 42 ( 6 ): 449 – 54 . Google Scholar CrossRef Search ADS PubMed 24. Dupont WD , Plummer WD Jr . Power and sample size calculations for studies involving linear regression . Control Clin Trials 1998 ; 19 : 589 – 601 . Google Scholar CrossRef Search ADS PubMed 25. Moher D , Hopewell S , Schulz KF , et al. ; Consolidated Standards of Reporting Trials Group . CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials . J Clin Epidemiol 2010 ; 63 : e1 – 37 . Google Scholar CrossRef Search ADS PubMed 26. Jackson A , Wilkinson C , Ranger M , Pill R , August P . Can primary prevention or selective screening for melanoma be more precisely targeted through general practice? A prospective study to validate a self administered risk score . BMJ 1998 ; 316 : 34 – 9 ; discussion 38–9. Google Scholar CrossRef Search ADS PubMed 27. Usher-Smith JA , Kassianos AP , Emery JD , et al. Identifying people at higher risk of melanoma across the U.K.: a primary-care-based electronic survey . Br J Dermatol 2017 ; 176 : 939 – 48 . Google Scholar CrossRef Search ADS PubMed 28. Rat C , Quereux G , Riviere C , et al. Targeted melanoma prevention intervention: a cluster randomized controlled trial . Ann Fam Med 2014 ; 12 : 21 – 8 . Google Scholar CrossRef Search ADS PubMed 29. Quéreux G , Moyse D , Lequeux Y , et al. Development of an individual score for melanoma risk . Eur J Cancer Prev 2011 ; 20 : 217 – 24 . Google Scholar CrossRef Search ADS PubMed 30. Bayram C , Valenti L , Britt H . General practice encounters with men . Aust Fam Physician 2016 ; 45 : 171 – 4 . Google Scholar PubMed 31. Hummers-Pradier E , Scheidt-Nave C , Martin H , et al. Simply no time? Barriers to GPs’ participation in primary health care research . Fam Pract 2008 ; 25 : 105 – 12 . Google Scholar CrossRef Search ADS PubMed 32. Rosemann T , Szecsenyi J . General practitioners’ attitudes towards research in primary care: qualitative results of a cross sectional study . BMC Fam Pract 2004 ; 5 : 31 . Google Scholar CrossRef Search ADS PubMed 33. Seité S , Del Marmol V , Moyal D , Friedman AJ . Public primary and secondary skin cancer prevention, perceptions and knowledge: an international cross-sectional survey . J Eur Acad Dermatol Venereol 2017 ; 31 : 815 – 20 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Family Practice Oxford University Press

Personalized melanoma risk assessments and tailored prevention advice: a pragmatic randomized controlled trial in Australian general practice

Loading next page...
 
/lp/ou_press/personalized-melanoma-risk-assessments-and-tailored-prevention-advice-eZOMI8Kmbs
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0263-2136
eISSN
1460-2229
D.O.I.
10.1093/fampra/cmy040
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Personalized risk assessments using prediction models that incorporate several melanoma risk factors may promote melanoma-prevention behaviours. Objectives To evaluate the effect on short-term melanoma-prevention behaviours of web-based, real-time, model-generated personalized melanoma risk information and tailored prevention advice, and its feasibility and clinician acceptability. Methods Between February and April 2016, in an open randomized controlled trial across four general medical practices in New South Wales, Australia, 272 patients were randomly allocated to receive (i) real-time model-generated personalized melanoma risk assessment and tailored melanoma-prevention advice or (ii) generic melanoma-prevention advice. We measured self-reported melanoma-prevention behaviours at baseline and 6 weeks and the intervention’s feasibility and acceptability. Results Follow-up questionnaires were completed by 185 patients at 6 weeks: 174 assessed as average risk and 11 as high or very high risk. There were no statistically significant differences between intervention and control patients in sun protection, sun exposure or early diagnosis behaviours. When stratified by melanoma risk, average risk patients in the intervention group appeared to show greater sun protection at 6 weeks (mean difference = 0.23, on a scale of 1–5; 95% confidence interval: 0.01 to 0.45; P = 0.04) than patients in the control group; the P value for interaction between intervention and risk category was 0.10. There was favourable feedback from patients and general practitioners. Conclusions Web-based delivery in general practice of real-time, model-generated personalized melanoma risk prediction and tailored melanoma-prevention advice is feasible and acceptable. An apparent increase in sun protection behaviour in average risk patients warrants further evaluation in different risk groups. Health promotion, melanoma, primary health care, primary prevention, risk reduction behaviour, secondary prevention Introduction Primary prevention interventions based on sun protection can reduce the melanoma burden (1,2). Despite community-wide mass media campaigns (3), sun protection levels remain relatively low (4–9). Risk prediction models, which provide a personalized estimate of risk based on a combination of melanoma risk factors, and tailoring prevention interventions to melanoma risk (10,11) may motivate greater adoption of sun protection behaviours. Few melanoma risk prediction models have undergone external validation in independent populations or prospective evaluation in routine clinical settings (12,13). Based on data from an Australian population–based case–control study, we have developed a melanoma risk prediction model of self-assessed melanoma risk factors: hair colour, naevus density, first-degree family history, previous non-melanoma skin cancer and lifetime sunbed use (14). External validation of our model showed good discrimination between those with and without melanoma (area under the receiver operating curve between 0.63 and 0.67 across four independent population-based studies), good calibration and higher net benefit than obtained by simply classifying everyone as high or low risk (14). We aimed to evaluate prospectively the effect on short-term melanoma-prevention behaviours of delivering real-time model-generated personalized melanoma risk assessments and tailored prevention advice in general practice, and to assess the feasibility and ‘clinician acceptability’ of using the model in practice. Methods Design, setting and participants We conducted an open, individually randomized controlled trial with 1:1 allocation of participants to one of two parallel groups in four general practices (three urban and one rural, comprising 29 general practitioners) in New South Wales, Australia, from 16 February to 30 April 2016. Consecutive patients were invited from the general practice waiting rooms on different days and at different times to participate. Eligible patients were aged over 18 years and had the capacity to give informed consent in English. We excluded patients with a previous diagnosis of cutaneous melanoma. Intervention Before they saw the general practitioner, all eligible consenting patients completed a baseline electronic questionnaire delivered using a tablet computer and a web-based application developed for this study (Supplementary Figure S1). The web-based application randomly allocated patients to the intervention or control group, and delivered to intervention group patients a real-time, on-screen and printed model-generated personalized melanoma risk assessment and tailored prevention advice. Patients in each group were given a booklet containing generic melanoma-prevention information on risk factors and prevention measures. The melanoma risk assessments and tailored prevention advice were not given to the doctors separately from the patients. The risk assessments were communicated as absolute remaining lifetime risk of melanoma (to 85 years old) with a 100-person pictograph, a relative remaining lifetime risk (relative to average remaining lifetime risk in people of the same 5-year age and sex stratum in New South Wales), and a risk category (average, high or very high) (Fig. 1). Since most Australians have fair skin and almost all epidemiological studies evaluating melanoma risk factors have been conducted in fair-skinned populations, patients at low melanoma risk were allocated to the average risk category. We defined the risk category cut-points and tailored the prevention advice to existing cut-points specified in the Australian guidelines for preventive activities in general practice, which are based on published relative risks for individual risk factors (15). This advice recommended sun protection for those at average risk (relative risk ≤2), sun protection and a full-body clinical examination for those at high risk (relative risk between 2 and 6), and sun protection, a full-body clinical examination and full-body skin self-examination for those at very high risk (relative risk ≥6). Figure 1. View largeDownload slide View largeDownload slide View largeDownload slide Model-generated melanoma risk assessments and tailored melanoma–prevention advice from February to April 2016 Figure 1. View largeDownload slide View largeDownload slide View largeDownload slide Model-generated melanoma risk assessments and tailored melanoma–prevention advice from February to April 2016 Outcomes The primary outcome was short-term sun protection behaviours—seeking shade, sunscreen use, wearing sunglasses, wearing a hat and protective clothing—which were individually measured and summarized in a validated composite measure (16). Secondary outcomes included sun exposure, full-body clinical- and skin self-examinations, behavioural intentions, melanoma risk perception, feasibility and clinician acceptability. Data collection Patients were contacted by their preferred method (email, post or telephone) 6 weeks after they had completed the baseline questionnaire. Non-responders were sent up to three reminders at 2-week intervals. The baseline questionnaire contained questions on sociodemographic characteristics, self-assessed melanoma risk factors in our melanoma risk prediction model (14) and validated questions on sun protection behaviours, sun exposure, clinical- and skin self-examinations, melanoma-prevention intentions and melanoma risk perception (16–18). The follow-up questionnaire contained the same questions (Supplementary Figure S1). Sun protection behaviours were measured on a five-point Likert scale with reference to hours between 9 am and 5 pm during the last week, ranging from 1 corresponding with ‘never’ to 5 corresponding with ‘always’ and were averaged to form a composite measure. Patients were also asked to indicate their usual time spent outdoors between 9 am and 5 pm during the last week, and whether they had a clinical examination and a skin self-examination during the last 12 months (baseline) or the last 6 weeks (follow-up). Melanoma-prevention intentions, based on the trans-theoretical model of behaviour change (19), were measured on a five-point Likert scale with reference to hours between 9 am and 5 pm during the last week, ranging from 1 corresponding with ‘I have never thought of doing this’ (pre-contemplation) to 5 corresponding with ‘I have been doing this for quite a while’ (maintenance) and were averaged to form a composite measure. Melanoma risk perception was also measured on a five-point Likert scale using the question ‘How would you rate your risk of developing melanoma compared to other men or women of your age?’; ranging from 1 ‘much lower’ to 5 ‘much higher’ than average risk. We measured the time taken to complete the baseline electronic questionnaire, and time taken for intervention group patients to review their personalized melanoma risk assessments and tailored prevention advice on the web-based application. Intervention group patients were asked to evaluate the web-based risk assessments (Supplementary Material 2). Participating general practitioners were asked to evaluate the intervention after the study by responding to a questionnaire (Supplementary Material 3). Sample size Our power calculations indicated that we needed to randomize 262 participants to detect a minimum difference between the control and intervention group of 0.2 in the composite sun protection behaviours scale (range: 1–5), assuming a standard deviation (SD) of 2.5, a 5% two-sided significance level, 80% power and 25% lost to follow-up (20–23). The sample sizes were determined using PS software (24). Randomization Patients were randomized using a computer-generated schedule with randomly varied block sizes within each 5-year age and sex stratum. Allocation was assigned by the web-based application and concealed. Patients were not informed of their group; however, blinding of the researchers was not possible. Statistical methods Descriptive statistics were calculated for all variables. Sun protection behaviour, sun exposure, melanoma-prevention intentions and melanoma risk perception scores were analysed as continuous variables using analysis of covariance adjusted for age, sex, recruitment site and corresponding baseline value. Clinical- and skin self-examinations (dichotomized to adherent versus non-adherent to guidelines for preventive activities (15) based on estimated personalized melanoma risk) were analysed using log-binomial models adjusted for age, sex, site of recruitment and corresponding baseline value. We compared the baseline characteristics between those who were lost to follow-up and those who completed the follow-up questionnaire. Data were analysed on an intention-to-treat basis using SAS 9.4 statistical software with a two-sided significance level of 0.05. We report methods and results in accordance with the Consolidated Standards of Reporting Trials statement (25). Results Participants We approached 411 general practice patients; 355 (86.3%) who agreed to participate were assessed for eligibility. Among the 320 (90.1%) patients who were eligible and gave consent, 48 (15.0%) were called in to see their doctor before completing the baseline electronic questionnaire; therefore, 272 patients were randomized (Fig. 2). The baseline sociodemographic characteristics and melanoma risk profiles were similar in the two groups (Table 1). Among the 134 intervention group patients, the mean (SD) absolute remaining lifetime melanoma risk was 3.11% (5.0%) and mean relative remaining lifetime risk was 0.90 (1.15). Among the 138 control group patients, the mean (SD) absolute remaining lifetime melanoma risk was 2.94% (4.73%) and the mean (SD) relative remaining lifetime risk was 0.80 (1.05). Table 1. Baseline characteristics for general practice patients who were offered generic and personalized melanoma risk information from February to April 2016 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 aPictogram shown in Supplementary Figure 1 (question 9). View Large Table 1. Baseline characteristics for general practice patients who were offered generic and personalized melanoma risk information from February to April 2016 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 Variable Control (n = 138) Intervention (n = 134) Age, mean (standard deviation) 45.20 (16.26) 45.75 (16.23) Sex  Male 40 37  Female 98 97 Education  High school 49 48  Trade certificate or diploma 33 15  University degree 75 91 Country of birth  Australia or New Zealand 110 100  United Kingdom 5 8  Other country 23 26 Self-reported health status  Excellent 14 15  Very good 41 49  Good 62 54  Poor 21 16 Natural hair colour  Red 2 6  Blonde 14 19  Brown 53 55  Black 69 54 Naevus densitya  None 29 32  Few 80 73  Some 24 26  Many 5 3 Melanoma family history (yes) 29 23 Non-melanoma skin cancer (yes) 14 15 Sunbed use  None 112 113  1–10 sessions 20 13  ≥10 sessions 6 8 Risk group  Average risk 132 125  High risk 5 8  Very high risk 1 1 aPictogram shown in Supplementary Figure 1 (question 9). View Large Figure 2. View largeDownload slide Flow of participants from February to April 2016 through each stage of the randomized controlled trial Figure 2. View largeDownload slide Flow of participants from February to April 2016 through each stage of the randomized controlled trial Outcomes Melanoma-prevention behaviours, intentions and risk perception Six-week follow-up questionnaires were completed by 185 (68.0%) patients, 174 (94.1%) assessed as average melanoma risk and 11 (5.9%) as high or very high melanoma risk. There were no overall significant differences at follow-up between intervention and control patients in sun protection (P = 0.13), sun exposure (P = 0.82 for weekday exposure and P = 0.33 for weekend exposure), early diagnosis behaviours (P = 0.98 for clinical examination and P = 0.09 for skin self-examination), melanoma-prevention intentions or melanoma risk perception (Table 2). There were no new full-body clinical examinations and six new full-body skin self-examinations (five intervention group patients assessed as average melanoma risk and one control group patient assessed as very high melanoma risk). When stratified by risk category, sun protection behaviours (composite score) in those at average risk of melanoma were higher in intervention than control patients [mean difference = 0.23, on a scale of 1–5; 95% confidence interval (CI): 0.01 to 0.45; P = 0.04], as was use of sunglasses (mean difference = 0.43, on a scale of 1–5; 95% CI: 0 to 0.86; P = 0.05). The P values for the interactions of intervention by risk group were P = 0.10 for sun protection and P = 0.50 for sunglasses; however, the study was underpowered to detect meaningful interactions between risk groups and sun protection behaviours. Table 2. Differences between general practice patients who were offered personalized or generic melanoma risk information at 6-week follow-up from February to April 2016, adjusting for age, sex, site of recruitment and corresponding baseline value Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA NA, not applicable; SD, standard deviation. aRange of values was 1–5 (1 = never and 5 = always). bThe Likert scale responses were averaged to form a composite scale. cRange of values was 1–5 (1 = I have never thought of doing this and 5 = I have been doing this for quite a while). dRange of values was 1–5 (1 = much lower than average and 5 = much higher than average). eRange of values was 1–5 (1 = not at all concerned and 5 = very concerned). fAt baseline, patients were asked whether they had a skin examination in the last 12 months. At follow-up, the question referred to the last 6 weeks only. View Large Table 2. Differences between general practice patients who were offered personalized or generic melanoma risk information at 6-week follow-up from February to April 2016, adjusting for age, sex, site of recruitment and corresponding baseline value Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA Baseline Follow-up after 6 weeks Overall (n = 272) Overall (n = 185) Average risk (n = 174) High or very high risk (n = 11) Generic risk group at baseline (n = 138) Personalized risk group at baseline (n = 134) Generic risk group at follow-up (n = 96) Personalized risk group at follow-up (n = 89) Difference (intervention–control) between groups at follow-up Generic risk group at follow-up (n = 92) Personalized risk group at follow-up (n = 82) Difference (intervention–control) between groups at follow-up Generic risk group at follow- up (n = 4) Personalized risk group at follow-up (n = 7) Difference (intervention–control) between groups at follow-up Variables Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value Mean (SD) Mean (SD) Mean difference (95% confidence interval) P Value P Value for interaction between subgroup and intervention group Sun protection behavioursa Seek shade 3.86 (0.72) 3.66 (0.79) 3.95 (0.10) 3.82 (0.09) –0.13 (–0.35 to 0.09) 0.24 3.98 (0.92) 3.89 (0.84) –0.09 (–0.30 to 0.13) 0.42 2.99 (0.51) 2.81 (0.52) –0.18 (–0.95 to 0.58) 0.54 0.05 Wear sunscreen 2.86 (1.52) 2.54 (1.50) 2.44 (0.20) 2.64 (0.18) 0.20 (–0.23 to 0.62) 0.36 2.39 (1.91) 2.73 (1.71) 0.34 (–0.10 to 0.78) 0.13 2.92 (0.87) 1.69 (0.77) –1.23 (–2.45 to –0.01) 0.05 0.02 Wear sunglasses 3.89 (1.41) 3.42 (1.59) 3.23 (0.19) 3.57 (0.18) 0.34 (–0.08 to 0.75) 0.11 3.11 (1.83) 3.54 (1.64) 0.43 (0.002 to 0.86) 0.05 5.76 (1.69) 3.88 (1.53) –1.88 (–4.30 to 0.55) 0.10 0.50 Wear hat 2.84 (1.47) 2.53 (1.45) 2.64 (0.18) 2.74 (0.17) 0.10 (–0.29 to 0.49) 0.61 2.56 (1.75) 2.66 (1.57) 0.11 (–0.30 to 0.51) 0.61 4.08 (1.19) 3.47 (1.11) –0.60 (–2.26 to 1.05) 0.37 0.57 Wear sun protective clothing 2.60 (1.39) 2.55 (1.44) 2.39 (0.17) 2.68 (0.16) 0.29 (–0.09 to 0.66) 0.13 2.34 (1.65) 2.63 (1.48) 0.29 (–0.09 to 0.67) 0.13 3.23 (2.06) 3.27 (1.69) 0.05 (–2.86 to 2.95) 0.97 0.62 Compositeb 3.21 (0.80) 2.94 (0.84) 2.92 (0.10) 3.08 (0.09) 0.16 (–0.05 to 0.38) 0.13 2.86 (1.29) 3.09 (1.23) 0.23 (0.01 to 0.45) 0.04 3.83 (0.67) 3.01 (0.59) –0.81 (–1.74 to 0.11) 0.07 0.10 Sun exposure (self-reported average per day in minutes) Weekday 127.71 (105.93) 130.86 (102.97) 137.28 (12.97) 133.92 (12.18) –3.37 (–31.75 to 25.02) 0.82 132.75 (126.93) 132.22 (114.74) –0.53 (–30.07 to 29.00) 0.97 288.59 (186.42) 223.24 (150.02) –65.35 (–222.23 to 91.63) 0.31 0.83 Weekend 181.56 (121.57) 190.73 (116.39) 156.15 (20.29) 178.07 (19.25) 21.92 (–22.35 to 66.19) 0.33 159.59 (200.85) 176.51 (184.55) 16.92 (–29.74 to 63.58) 0.48 49.43 (204.45) 154.78 (177.48) 105.35 (–180.09 to 390.78) 0.36 0.53 Intentions to undertake sun protectionc Intention to seek shade 4.19 (1.32) 4.32 (1.28) 3.86 (0.18) 3.83 (0.16) –0.03 (–0.41 to 0.36) 0.90 3.88 (1.76) 3.91 (1.57) 0.03 (–0.38 to 0.43) 0.89 3.19 (1.55) 3.15 (1.27) –0.04 (–2.21 to 2.13) 0.96 0.30 Intention to wear sunscreen 3.95 (1.40) 4.01 (1.40) 3.97 (0.16) 3.74 (0.15) –0.24 (–0.58 to 0.11) 0.18 3.97 (1.55) 3.85 (1.39) –0.12 (–0.47 to 0.24) 0.52 4.29 (1.44) 2.80 (1.40) –1.49 (–3.57 to 0.59) 0.12 0.06 Intention to wear sunglasses 4.14 (1.39) 4.13 (1.37) 4.28 (0.12) 4.38 (0.11) 0.10 (–0.16 to 0.36) 0.43 4.26 (1.21) 4.38 (1.09) 0.12 (–0.15 to 0.40) 0.38 4.29 (0.94) 4.34 (0.70) 0.05 (–1.28 to 1.38) 0.92 0.81 Intention to wear hat 3.53 (1.45) 3.76 (1.40) 3.79 (0.15) 3.66 (0.14) –0.12 (–0.44 to 0.19) 0.44 3.77 (1.44) 3.67 (1.30) –0.10 (–0.43 to 0.23) 0.56 3.96 (1.06) 3.51 (1.08) –0.45 (–1.97 to 1.07) 0.46 0.97 Intention to wear sun protective clothing 2.87 (1.62) 3.23 (1.69) 3.15 (0.17) 3.31 (0.16) 0.16 (–0.22 to 0.54) 0.41 3.11 (1.72) 3.28 (1.53) 0.17 (–0.23 to 0.57) 0.40 3.99 (0.92) 3.62 (0.80) –0.37 (–1.67 to 0.93) 0.47 0.79 Compositeb 3.74 (1.00) 3.89 (0.96) 3.83 (0.10) 3.78 (0.09) –0.04 (–0.27 to 0.18) 0.70 3.81 (1.41) 3.82 (1.32) 0.00 (–0.23 to 0.24) 0.99 4.09 (0.59) 3.47 (0.52) –0.61 (–1.44 to 0.21) 0.11 0.42 Intention to undertake early detection behavioursc Intention to undertake a clinical examination 3.06 (1.61) 2.97 (1.65) 3.71 (0.13) 3.83 (0.13) 0.12 (–0.16 to 0.41) 0.40 3.65 (1.34) 3.77 (1.23) 0.11 (–0.19 to 0.42) 0.46 4.74 (0.75) 4.38 (0.59) –0.36 (–1.42 to 0.69) 0.39 0.58 Intention to undertake a skin self-examination 2.91 (1.66) 2.79 (1.63) 3.52 (0.16) 3.54 (0.15) 0.02 (–0.32 to 0.36) 0.90 3.47 (1.57) 3.46 (1.42) –0.01 (–0.37 to 0.35) 0.96 4.25 (1.21) 4.43 (1.04) 0.18 (–1.48 to 1.83) 0.78 0.42 Melanoma risk perceptiond 2.73 (0.87) 2.78 (0.86) 2.69 (0.10) 2.75 (0.09) 0.07 (–0.15 to 0.28) 0.55 2.63 (0.94) 2.63 (0.84) 0.002 (–0.22 to 0.21) 0.98 4.01 (0.85) 4.20 (0.66) 0.19 (–1.04 to 1.41) 0.69 0.00 Melanoma concerne 2.35 (0.96) 2.65 (1.22) 2.57 (0.12) 2.55 (0.11) –0.01 (–0.27 to 0.24) 0.91 2.52 (1.07) 2.51 (0.96) –0.02 (–0.27 to 0.23) 0.89 3.81 (1.67) 3.62 (1.60) –0.19 (–2.69 to 2.32) 0.85 0.47 Early diagnosis behaviours (% adherent to guidelines based on melanoma risk)f n (%) n (%) n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value n (%) n (%) Relative risk (95% confidence interval) P Value Clinical examination 97 (70.29) 102 (76.12) 96 (100) 82 (92.13) 0.33 (0.10 to 1.16) 0.08 92 (100) 82 (100) NA NA 4 (100) 7 (100) NA NA NA Skin self-examination 119 (86.23) 119 (88.81) 96 (100) 83 (93.26) 0.09 (0.008 to 1.13) 0.06 92 (100) 77 (93.90) 0.15 (0.02 to 1.46) 0.10 4(100) 6 (85.71) 1.06 (0.03 to 35.84) 0.97 NA NA, not applicable; SD, standard deviation. aRange of values was 1–5 (1 = never and 5 = always). bThe Likert scale responses were averaged to form a composite scale. cRange of values was 1–5 (1 = I have never thought of doing this and 5 = I have been doing this for quite a while). dRange of values was 1–5 (1 = much lower than average and 5 = much higher than average). eRange of values was 1–5 (1 = not at all concerned and 5 = very concerned). fAt baseline, patients were asked whether they had a skin examination in the last 12 months. At follow-up, the question referred to the last 6 weeks only. View Large Feasibility Patients took on average 5 minutes and 14 seconds to complete the baseline electronic questionnaire. Intervention group patients took on average 1 minute and 35 seconds to review their personalized melanoma risk assessments and 47 seconds to review their tailored prevention advice on the web-based application. Intervention patients reported the web-based melanoma risk assessment as easy to use (96%), easy to understand (97%) and useful (90%). Clinician acceptability We mailed a short questionnaire to 29 general practitioners who worked in the participating general practices; six (21%, two men and four women, aged 31–62 years old) completed it. All six general practitioners reported that the melanoma risk assessments and tailored prevention advice generated by the web-based application were easy to understand and unlikely to make the patients feel upset or worried. Five (83%) said they would be likely to use the web-based application if it became widely available. Conclusions The delivery of real-time model-generated personalized melanoma risk and tailored prevention advice in general practice using a web-based melanoma risk prediction model is highly feasible and acceptable. Although we did not find any overall differences between all intervention and control patients in sun protection, sun exposure, early diagnosis behaviours, melanoma-prevention intentions or risk perception at 6-week follow-up, there were modest increases in sun protection behaviours in intervention patients compared with control patients among those at average melanoma risk. We could find only three previously reported instances of use of melanoma risk prediction models in general practice (26–28). One of the studies (Rat et al. 2014) (28), conducted in western France, appeared similar to ours. It evaluated the effect of model-generated melanoma risk assessments (29) on sun protection behaviours in a clustered-randomized controlled trial where general practitioners were randomized to deliver targeted screening and education intervention or conventional information-based campaign to 173 patients identified as at high risk. At 5-month follow-up, high-risk patients in the intervention group were less likely to sunbathe and more likely to have performed skin self-examinations compared with the high-risk patients in the control group (28). It is not clear in this study whether patients in the intervention group were aware of their high risk, as they were in our study, and, if they were, whether this knowledge contributed to their more favourable outcomes. The risk category classifications and proportion of patients classified as high risk were quite different between the study of Rat et al., which classified high risk using a model relative risk category cut-point of 11 and 46% of patients as at high risk, and our study, which used relative risk category cut-points from the Australian guidelines for preventive activities in general practice (15) and classified 6% of patients as high or very high risk. It is worth noting that the risk category cut-points given in the Australian guidelines were based on individual risk factors, not on a combined risk factor model. If we instead classified our participants based on their individual risk factors using the same relative risk cut-points (2 for high and 6 for very high risk), we would have classified 19% of patients as at high or very high risk. Choosing appropriate cut-points to define levels of risk is not straightforward and depends on an assessment of the net benefit of different cut-points and the relative benefits and harms of the intervention (14). These issues need further investigation before they can be readily used to inform the choice of cut-points for classifying high risk. While the approach we used was objective, it was a disadvantage that the small numbers of patients classified as high or very high risk substantially diminished its power to evaluate interaction by subgroup. To our knowledge, ours is the first study to deliver real-time model-generated personalized melanoma risk and tailored prevention advice to patients across all risk profiles. Its main strengths are the randomized controlled trial design and the pragmatic evaluation in routine general practice, in which 86% of approached patients agreed to participate. The study also included practices from both urban and rural areas to represent the breadth of the population. Furthermore, we used a novel method of data collection and risk profile delivery, using a web-based application on a tablet computer to minimize time taken and transcription errors. Randomization was also facilitated within the application using a custom-built computer algorithm. It was a potential limitation that we recruited more women than men; however, this reflects the general practice consulting population (30). While the general practitioner participation in evaluating the intervention was low, it was in line with other studies (31,32). Thirty-two per cent of patients in the randomized controlled trial were lost to follow-up, which might have introduced selection bias. We were not able to collect information from non-responders, but those who were lost to follow-up had similar baseline characteristics to those who completed the follow-up questionnaire (Supplementary Material 4). We recruited participants from summer to early autumn and completed the follow-up in winter when solar ultraviolet radiation is lower; however, while season and weather influence sun protection behaviours, changes in season would have affected the control and intervention groups equally. We delivered the intervention directly to patients and did not measure whether the patients discussed the melanoma risk information and prevention advice with their general practitioners, which would require an audit of the medical records. We also did not measure melanoma incidence or mortality as outcomes; which would require a much larger study with very long follow-up. Tailored prevention advice based on real-time model-generated risk assessment in general practice is highly feasible and acceptable and suggests that similar approaches to offer prevention, screening or support that is tailored to the characteristics and needs of the patient could be used more widely for other diseases in primary and secondary care settings in Australia and other countries. Low melanoma-prevention behaviours were observed in a recent international cross-sectional survey conducted across 23 countries, with higher prevention behaviours among people from countries with higher ambient ultraviolet radiation like Australia, Chile and Greece (33). Our findings provide some evidence for increased short-term sun protection behaviours among those at average melanoma risk who received melanoma risk prediction and tailored prevention advice. Future studies should be large enough to analyse within each risk group and probably use a lower absolute risk category cut-point to define greater numbers of patients as at high risk. Meanwhile, more research is needed to establish a broadly applicable, objective basis for defining high risk when using risk prediction models in preventive interventions. Supplementary Material Supplementary data are available at Family Practice online. Declaration Funding: KV was supported by a University of Sydney Postgraduate Scholarship in Cancer Epidemiology (funded through AEC’s Cancer Institute NSW fellowship), a Sydney Catalyst Top-Up Research Scholar Award, and a Primary Care Collaboration Cancer Clinical Trials Group and Clinical Oncology Society of Australia Training Award in Cancer and Primary Care. AEC was supported by fellowships from the Cancer Institute NSW (15/CDF/1–14) and the National Health and Medical Research Council (NHMRC) (1147843). The funding organizations had no role in the study design, data collection, analysis or interpretation of data, and writing or submitting the manuscript. Ethical approval: The study was approved by The University of Sydney Human Research Ethics Committee (2014/144) and prospectively registered with the Australian New Zealand Clinical Trials Registry (ACTRN12615001019594) available at https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?ACTRN=12615001019594. Conflict of interest: None declared. Acknowledgements We wish to thank patients and staff at Montrose Medical Practice, Gordon Family Medical Practice, Summer Hill Village Medical Practice and Sussex Inlet Medical Centre for their participation. We also wish to thank Allison Grech, Brooke Beswick and Amelia Smit for their help with recruitment and data management. References 1. Armstrong BK , Kricker A . How much melanoma is caused by sun exposure ? Melanoma Res 1993 ; 3 : 395 – 401 . Google Scholar CrossRef Search ADS PubMed 2. Olsen CM , Wilson LF , Green AC , et al. Cancers in Australia attributable to exposure to solar ultraviolet radiation and prevented by regular sunscreen use . Aust N Z J Public Health 2015 ; 39 : 471 – 6 . Google Scholar CrossRef Search ADS PubMed 3. Wu YP , Aspinwall LG , Conn BM , et al. A systematic review of interventions to improve adherence to melanoma preventive behaviors for individuals at elevated risk . Prev Med 2016 ; 88 : 153 – 67 . Google Scholar CrossRef Search ADS PubMed 4. Montague M , Borland R , Sinclair C . Slip! Slop! Slap! and SunSmart, 1980–2000: skin cancer control and 20 years of population-based campaigning . Health Educ Behav 2001 ; 28 ( 3 ): 290 – 305 . Google Scholar CrossRef Search ADS PubMed 5. Sinclair C , Foley P . Skin cancer prevention in Australia . Br J Dermatol 2009 ; 161 ( suppl 3 ): 116 – 23 . Google Scholar CrossRef Search ADS PubMed 6. Dobbinson S , Wakefield M , Hill D , et al. Prevalence and determinants of Australian adolescents’ and adults’ weekend sun protection and sunburn, summer 2003–2004 . J Am Acad Dermatol 2008; 59 ( 4 ): 602 – 14 . CrossRef Search ADS PubMed 7. Dobbinson SJ , Wakefield MA , Jamsen KM , et al. Weekend sun protection and sunburn in Australia trends (1987–2002) and association with SunSmart television advertising . Am J Prev Med 2008 ; 34 : 94 – 101 . Google Scholar CrossRef Search ADS PubMed 8. Volkov A , Dobbinson S , Wakefield M , Slevin T . Seven-year trends in sun protection and sunburn among Australian adolescents and adults . Aust N Z J Public Health 2013 ; 37 : 63 – 9 . Google Scholar CrossRef Search ADS PubMed 9. Koch S , Pettigrew S , Minto C , et al. Trends in sun-protection behaviour in Australian adults 2007–2012 . Australas J Dermatol 2017 ; 58(2): 111–6. 10. Ahmed H , Naik G , Willoughby H , Edwards AG . Communicating risk . BMJ 2012 ; 344 : e3996 . Google Scholar CrossRef Search ADS PubMed 11. Freedman AN , Seminara D , Gail MH , et al. Cancer risk prediction models: a workshop on development, evaluation, and application . J Natl Cancer Inst 2005 ; 97 : 715 – 23 . Google Scholar CrossRef Search ADS PubMed 12. Vuong K , McGeechan K , Armstrong BK , Cust AE . Risk prediction models for incident primary cutaneous melanoma: a systematic review . JAMA Dermatol 2014 ; 150 : 434 – 44 . Google Scholar CrossRef Search ADS PubMed 13. Wernli KJ , Henrikson NB , Morrison CC , et al. Screening for skin cancer in adults: updated evidence report and systematic review for the US preventive services task force . JAMA 2016 ; 316 : 436 – 47 . Google Scholar CrossRef Search ADS PubMed 14. Vuong K , Armstrong BK , Weiderpass E , et al. ; Australian Melanoma Family Study Investigators . Development and external validation of a melanoma risk prediction model based on self-assessed risk factors . JAMA Dermatol 2016 ; 152 : 889 – 96 . Google Scholar CrossRef Search ADS PubMed 15. RACGP . Guidelines for Preventive Activities in General Practice . 9th edn . East Melbourne, Victoria : The Royal Australian College of General Practitioners , 2016 . 16. Glanz K , Yaroch AL , Dancel M , et al. Measures of sun exposure and sun protection practices for behavioral and epidemiologic research . Arch Dermatol 2008 ; 144 : 217 – 22 . Google Scholar CrossRef Search ADS PubMed 17. O’Riordan DL , Nehl E , Gies P , et al. Validity of covering-up sun-protection habits: association of observations and self-report . J Am Acad Dermatol 2009 ; 60 : 739 – 44 . Google Scholar CrossRef Search ADS PubMed 18. Bränström R , Kristjansson S , Ullén H , Brandberg Y . Stability of questionnaire items measuring behaviours, attitudes and stages of change related to sun exposure . Melanoma Res 2002 ; 12 : 513 – 9 . Google Scholar CrossRef Search ADS PubMed 19. Prochaska JO , Velicer WF . The transtheoretical model of health behavior change . Am J Health Promot 1997 ; 12 : 38 – 48 . Google Scholar CrossRef Search ADS PubMed 20. Robinson JK . Behavior modification obtained by sun protection education coupled with removal of a skin cancer . Arch Dermatol 1990 ; 126 : 477 – 81 . Google Scholar CrossRef Search ADS PubMed 21. Glanz K , Schoenfeld ER , Steffen A . A randomized trial of tailored skin cancer prevention messages for adults: project SCAPE . Am J Public Health 2010 ; 100 : 735 – 41 . Google Scholar CrossRef Search ADS PubMed 22. Falk M , Magnusson H . Sun protection advice mediated by the general practitioner: an effective way to achieve long-term change of behaviour and attitudes related to sun exposure . Scand J Prim Health Care 2011 ; 29 ( 3 ): 135 – 43 . Google Scholar CrossRef Search ADS PubMed 23. Glazebrook C , Garrud P , Avery A , Coupland C , Williams H . Impact of a multimedia intervention “Skinsafe” on patients’ knowledge and protective behaviors . Prev Med 2006 ; 42 ( 6 ): 449 – 54 . Google Scholar CrossRef Search ADS PubMed 24. Dupont WD , Plummer WD Jr . Power and sample size calculations for studies involving linear regression . Control Clin Trials 1998 ; 19 : 589 – 601 . Google Scholar CrossRef Search ADS PubMed 25. Moher D , Hopewell S , Schulz KF , et al. ; Consolidated Standards of Reporting Trials Group . CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials . J Clin Epidemiol 2010 ; 63 : e1 – 37 . Google Scholar CrossRef Search ADS PubMed 26. Jackson A , Wilkinson C , Ranger M , Pill R , August P . Can primary prevention or selective screening for melanoma be more precisely targeted through general practice? A prospective study to validate a self administered risk score . BMJ 1998 ; 316 : 34 – 9 ; discussion 38–9. Google Scholar CrossRef Search ADS PubMed 27. Usher-Smith JA , Kassianos AP , Emery JD , et al. Identifying people at higher risk of melanoma across the U.K.: a primary-care-based electronic survey . Br J Dermatol 2017 ; 176 : 939 – 48 . Google Scholar CrossRef Search ADS PubMed 28. Rat C , Quereux G , Riviere C , et al. Targeted melanoma prevention intervention: a cluster randomized controlled trial . Ann Fam Med 2014 ; 12 : 21 – 8 . Google Scholar CrossRef Search ADS PubMed 29. Quéreux G , Moyse D , Lequeux Y , et al. Development of an individual score for melanoma risk . Eur J Cancer Prev 2011 ; 20 : 217 – 24 . Google Scholar CrossRef Search ADS PubMed 30. Bayram C , Valenti L , Britt H . General practice encounters with men . Aust Fam Physician 2016 ; 45 : 171 – 4 . Google Scholar PubMed 31. Hummers-Pradier E , Scheidt-Nave C , Martin H , et al. Simply no time? Barriers to GPs’ participation in primary health care research . Fam Pract 2008 ; 25 : 105 – 12 . Google Scholar CrossRef Search ADS PubMed 32. Rosemann T , Szecsenyi J . General practitioners’ attitudes towards research in primary care: qualitative results of a cross sectional study . BMC Fam Pract 2004 ; 5 : 31 . Google Scholar CrossRef Search ADS PubMed 33. Seité S , Del Marmol V , Moyal D , Friedman AJ . Public primary and secondary skin cancer prevention, perceptions and knowledge: an international cross-sectional survey . J Eur Acad Dermatol Venereol 2017 ; 31 : 815 – 20 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Family PracticeOxford University Press

Published: May 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off