Australia has among the highest incidence rate for melanoma in the world and in Australia, melanoma is the fourth most common cancer (excluding non-melanoma skin cancer). Its prevention is therefore a priority in this country. Sun exposure, particularly during childhood, is an important factor in the development of melanoma. Preventive efforts are therefore likely to be most effective if they involve young children. Although programs to encourage sun protection in Australian children exist, levels of sun exposure remain high.
The Kidskin study aimed to design, implement and evaluate an intervention to improve sun protection behaviour in young children. Kidskin was a non-randomised, control trial involving a cohort of 1,776 children who commenced school in 1995 at 5-6 years of age. The unit of intervention was the school and there were three groups: a “high intervention” group (eight schools), a “moderate intervention” group and a comparison group (14 schools). Comparison schools taught the standard WA health curriculum, while intervention schools received a specially designed sun protection curriculum. The high intervention group also received Kidskin program materials over the summer school holidays when sun exposure was likely to be highest, and were offered low cost sun protective swimwear at the start of summer each year. The interventions were delivered over four years from 1995 to 1999 when the children were in Years 1-4 at school.
Outcome measures included: number of melanocytic naevi (moles) on the back, chest (boys only), face and arms; tanning on the forearms and back; constitutional skin reflectance (inner arm); freckling on the face and arms, and parent report of child’s sun-related behaviour. Outcome data were collected at baseline in 1995 and at two post-tests in 1997 and 1999. Process evaluation data were collected each year from 1995-1999 and included teacher and parent use of and satisfaction with the classroom and home intervention. In 1995 and 1998 video recordings of the school playground were taken to measure children’s hat wearing and UV sensitive badges were pinned to children’s clothing at lunchtime to measure sun exposure during this period at school.
In 1999, after four years of intervention, intervention group children had approximately 5% fewer moles on their backs and between 3 and 11% fewer moles on other body sites relative to the comparison group (after adjustment for baseline naevus counts and potential confounders). However, these differences were not statistically significant and the high intervention was no more protective than the moderate intervention on moles.
After the first two years of the program, children in the intervention groups were less tanned at the end of summer than comparison group children. High intervention group children had the least suntan on their backs and forearms while comparison group children had the most. The results for the moderate intervention group were between the other two groups. By 1999, after four years of the Kidskin program, this effect was no longer evident and there was little or no difference between the study groups in terms of level of suntan.
The Kidskin intervention appears to have had positive effects on children’s sun protective behaviours as reported by their parents. In both years of follow up (1997 and 1999) a consistent gradient in results was seen across groups, with the high intervention groups having the most favourable behaviour, the comparison group the least favourable and the moderate intervention group in between. The beneficial effect of the Kidskin intervention was seen for time spent outdoors in the middle of the day, back coverage (using clothing and swimsuit styles that cover the trunk and upper arms) and shade use. There was little difference between groups in the use of sunscreen or wearing of hats.
Results showed that in some high intervention group schools the proportion of children wearing broad-brimmed or legionnaire-style hats was higher in 1998 than 1995. However, there was no change in the proportion wearing these hat types in either moderate or comparison group schools. Results also indicated no difference between the three groups in terms of lunchtime sun exposure at school. Therefore, the program seemed to have some effect on hat wearing in the playground, but did not change shade use by children at school during lunchtime.
Teachers reported they enjoyed teaching the program and taught between 66% and 79% of the Kidskin materials each year. High program dose levels during Years 1 to 3 were associated with increased back coverage, shade use and sunscreen use on the face and arms at post-test in 1999 compared to a low program dose. In the first two years of the program a high program dose was also related to increased back coverage and shade use compared to a medium program dose.
High program dose levels in the first three years of the program was related to reduced tanning on the forearm at post-test compared to a low program dose. A medium program dose in Year 1 and Years 1 to 4 was related to fewer naevi on the face and chest respectively, than a low program dose. A high cumulative program dose in the first three years of the intervention was associated with fewer naevi on the arms at post-test than a low dose.
Therefore, higher levels of program dose resulted in more positive outcomes, particularly related to sun protection of the arms. Further dissemination should encourage teaching of the program over the full school year to maximise implementation, particularly in the junior primary years when the program appeared most effective.
National Health and Medical Research Council
Cancer Foundation of Western Australia
The University of Western Australia, Professor Dallas English
Curtin University, Associate Professor Donna Cross
The University of Western Australia, Associate Professor Billie Giles-Corti
The University of Western Australia, Ms Elizabeth Milne
The University of Western Australia, Ms Christine Costa
The University of Western Australia, Ms Tommy Cordin
Curtin University, Ms Robyn Johnston