Photodermatology, Photoimmunology & Photomedicine

ORIGINAL ARTICLE

Self-reported sun-related knowledge, attitudes and behaviours among schoolchildren attending South African primary schools Caradee Y. Wright1, Patricia N. Albers2, Maria A. Oosthuizen1 & Nelvia Phala1

1

Council for Scientific and Industrial Research, Climate Studies, Modelling and Environmental Health Research Group, Pretoria, South Africa. 2 South African Medical Research Council, Environment and Health Research Unit, Houghton, South Africa.

SUMMARY Background Skin cancer and other adverse health effects result from excess solar ultraviolet radiation exposure. Sun protective practices are important interventions for skin cancer prevention, particularly when implemented early in life. Several international studies have assessed children’s sun-related knowledge, attitudes and behaviours in school settings but never before in Southern Africa, where multiethnic populations exist.

Key words: attitude; behaviour; knowledge; schoolchildren; South Africa; sun exposure

Purpose The purpose of this study is to describe the sun-related knowledge, attitudes and behaviours as self-reported by South African primary schoolchildren and consider the roles of sex and skin type.

Correspondence: Dr Caradee Yael Wright, PhD, Council for Scientific and Industrial Research, Climate Studies, Modelling and Environmental Health Research Group, PO Box 395, Pretoria, 0001, South Africa. Tel: +271 2841 3092 Fax: +271 2841 4257 e-mail: [email protected]

Accepted for publication: 20 December 2013

Conflicts of interest: None declared.

266

Methods A randomly selected sample of 707 schoolchildren from 24 government, urban schools in all nine provinces of South Africa were surveyed regarding their sun-related knowledge, attitudes and behaviours. Results Approximately 56% of students reported experience of sunburn last summer and 59% stated that they had got a suntan. Many students (64.5%) believed that one could protect oneself from getting skin cancer by avoiding getting sunburnt. Other means reported to do so by the students were to use sunscreen (65.4%), stay out of the summer sun (48.0%), cover up with clothing (45.5%) and eat the right foods (38.0%). Only about a quarter of the students (22.4%) wrongly agreed that it is safe to get sunburnt once or twice a year. Few students (8.7%) agreed that they like to have a suntan because they feel healthier and agreed that they think a suntan makes them feel more attractive to others (17.3%). Few also agreed that most of their friends (16.1%) and family (14.2%) think that a suntan is a good thing. Children reporting to have white/light brown skin (69.4%) were more likely to agree that they used sunscreen to protect themselves from getting sunburnt compared with children having brown/dark brown/black skin (54.8%) (P = 0.0005).

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd doi:10.1111/phpp.12107

Sun exposure among South African children

Conclusion South African schoolchildren at urban government schools do have some knowledge about sun protection, and they do have some positive sun behaviours; however, the reported occurrence of sunburn, a risk factor for skin cancer, was relatively high. There were few differences in responses by sex and some differences by skin type. These findings are important for the development of appropriate sun protection programmes aimed at schoolchildren in South Africa and other countries with similar multi-ethnic populations. Photodermatol Photoimmunol Photomed 2014; 30: 266–276

Some sun exposure is important for human health; however, excess sun exposure results in several adverse human health effects including eye diseases, immune suppression, sunburn and skin cancer. Several international epidemiological studies have shown that skin cancer is associated with acute sun exposure during childhood and adolescence (1). Children are at potential risk of sunburn during school hours when meal break times are scheduled during peak solar ultraviolet radiation (UVR) hours, i.e. between 10:00 h and 14:00 h. Despite this, sunburn and skin cancer are largely preventable with appropriate behavioural responses, i.e. reduced sun exposure during high solar UVR hours and the use of sun protection such as sunscreen and clothing. The Community Preventive Services Task Force found that sun protection awareness and intervention programmes implemented in primary schools are effective and can increase sun-protective behaviours, decrease sun exposure, sunburn incidence and formation of new moles and thereby reduce the risk of adverse health effects from appearing later in life (2, 3). Knowledge and attitudes influence sun protection behaviour in children and adolescents (4). A previous study carried out among New Zealand schoolchildren found that the relationship between knowledge and behaviours was mediated by attitudes (5). A recent Turkish study found that primary school children (11–14 years) did not practise sun protection behaviour despite knowledge being present about the dangers of the sun and how best to protect themselves (6). Some studies have found statistically significant differences in sun-related knowledge and attitudes for males and females (7), whereas other studies have not (5, 6). Hence, it is critical to determine sun-related knowledge, attitudes and behaviours for a specific population as differences have been found in different subpopulations, in order to tailor interventions for specific subpopulations and ultimately reduce adverse health effects overall.

In South Africa, approximately 850 deaths are attributed to melanoma skin cancer annually (8). The 2001 agestandardized incidence rate of melanoma in South Africa is approximately 22.5 and 17.4 per 100 000 for white males and white females, respectively (8). South Africa experiences relatively high levels of summertime erythemal (sunburning) solar UVR (9) with highest levels around midday. Despite these circumstances, until now, sun-related knowledge, attitudes and behaviours have not been recorded for a sample of South African primary schoolchildren. In South Africa, the Cancer Association of South Africa (CANSA) implements a SunSmart Awareness Programme that includes visits to schools and dissemination of sun protection informational materials. In 2010, CANSA produced the SunSmart Guide for Schools, which was made available on their web site. No SunSmart School Accreditation Programme presently exists in South Africa. Hence, it was decided to carry out a baseline study of South African schoolchildren’s sun-related knowledge, attitudes and behaviours to inform future efforts to increase targeted sun awareness campaigns in the country. This is the first study to do so and therefore provides important evidence to support the planning and development of skin cancer prevention campaigns for South Africa.

Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

267

METHODS Study design A cross-sectional, descriptive epidemiological study was applied to determine schoolchildren’s sun-related knowledge, attitudes and behaviours using self-reported questionnaires. The methodology was informed by the Haddon

Wright et al.

matrix and the Theory of Planned Behaviour. The study took place during the third school term of South Africa between August and October 2012 (late winter, early spring).

Sample The schools invited to participate in the research were randomly selected from the Department of Basic Education schools database. Schools were eligible for inclusion if they had classes of grade 7 learners, were public (government) urban (urban schools were chosen due to challenges in accessing rural schools) schools and where English or Afrikaans (being the two main languages in South Africa) was the main spoken and written language. Private schools, correspondence and home schools, and schools for children with special needs were excluded (because they require an alternative research methodology for inclusion). Schools with classes of Grade 7 students with fewer than 10 learners were excluded to optimize the use of the limited fieldwork budget. As the student questionnaires were posted to the schools, a random selection of government urban primary schools from all nine provinces was made to ensure fair representation of schools. The Department of Basic Education schools database was manually edited to exclude all ineligible schools identified above. From this edited list, 36 schools (four from each province) were randomly selected. School principals were telephoned, briefed about the study and invited to participate. When a school chose not to participate, the next randomly selected school from the same province was contacted and invited until the total of 36 schools was reached. The grade 7 schoolteacher was asked to give the grade 7 class learners 20 minutes of class time or break time to answer the questionnaire to minimize questionnaire loss and better the chance of retrieval. Each learner was provided with a brief project summary and a consent letter for their parent/ guardian/caregiver to sign to allow their child to complete the questionnaire. Children signed an assent form at the start of the questionnaire. The completed learner questionnaires were returned to the research team via pre-paid courier bags. Follow-up for questionnaire return was made when necessary, and a personalized thank you letter with a summary report was written and sent to the participating school principal, schoolteacher and grade 7 class.

Questionnaire The questionnaire was based on an instrument developed by the Anti-Cancer Council of Victoria during the 1990s that included items used in the South Australian Health 268

Omnibus surveys, the South Australian Triennial Sun Exposure Survey and the 2003–4 National Sun Protection Survey as well as similar instruments that have been developed for use among children in other countries (5, 7, 10, 11) but adjusted to local conditions, cultural differences and observed or reported sun-related behaviours specific to South Africa following piloting. Questions focused on (a) attitudes towards sun exposure, sun tan and sun protection; (b) knowledge of beneficial and harmful effects of the sun and sun protection; (c) behaviour when outdoors, i.e. hat use, sunscreen use, clothing for sun protection, sunburn experience, sun tanning experience, etc. and (d) demographic data, i.e. sex, age, skin type, etc. The questionnaire was pre-tested, translated into Afrikaans and piloted to achieve optimum comprehension and acceptability.

Ethical considerations and approvals Ethical approval for this study was obtained from the Council for Scientific and Industrial Research Ethics Committee (35/2012) on the 27 June 2012. Provincial approvals were granted by the following departments: KwaZulu-Natal Department of Education; Northern Cape Department of Education; Eastern Cape Department of Education; Department of Education: Mpumalanga; Western Cape Department of Education; Department of Education Free State; North-West Department of Basic Education; Gauteng Department of Education; and Limpopo Department of Education. School principals gave informed consent for the study prior to contacting the children. Children’s parents/guardians/caregivers completed an informed consent form, and children gave assent prior to completing the questionnaire.

Statistical analyses All questionnaire data were coded and entered in an electronic database. These data were cleaned and prepared for analysis in Microsoft Excel (2010). These data were then imported into Stata 11.0 (Stata Corp. 2009. Stata Statistical Software: Release II. College Station, TX: StataCorp LP) statistical analysis software for analyses, including descriptive statistics and bivariate analyses. Summary descriptive statistics included observed frequencies for all variables included in the questionnaire. Bivariate analysis included examining all variables by reported sex and skin type. Statistical significance was assessed using a P-value of less than or equal to 0.05; these were obtained using the chi-squared test, or in cases where cells had less than five observations, the Fisher’s Exact test was used. Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Sun exposure among South African children

Table 1. Sample descriptive statistics by province: schools and students

Province

Number of Percentage of total Number of participat- Percentage of schools invited Number of schools schools participated ing students per total students (n) participated (n) (%) participating school participating (%)

Gauteng 4 KwaZulu-Natal 4 Limpopo 4 Eastern Cape 4 Western Cape 4 Northern Cape 4 Mpumalanga 4 North-West 4 Free State 4 Total 36

2 2 3 4 2 2 3 3 3 24

8.3 8.3 12.5 16.8 8.3 8.3 12.5 12.5 12.5 100

RESULTS A total of 24 schools out of 36 schools (66.7% response rate, 12 schools did not complete the survey after agreeing to participate) who agreed to participate in the study, participated in the study and at least one of their grade 7 classes completed the questionnaire (Table 1). Reasons given for non-participation by schools were overcommitment to other research projects, lack of interest and lack of time. A total of 707 students completed the questionnaire; 38.1% were male, 61.4% were female and 4 respondents did not identify their sex (Table 2). As the sample is relatively small, no attempt was made to represent the country or provinces with these results; therefore, no weightings were applied. Most students (60.7%) were 13 years of age at the time of completing the questionnaire (0.8% were 11 years, 24.5% were 12 years and 13.7% were older than 13 years of age). The majority of students reported that they were black (39.8%), followed by white (26.7%), coloured (22.6%) (mixed European and African or Asian ancestry) and Indian/Asian (7.5%). From data obtained from Statistics South Africa for the 2011 census (available online at http://www.statssa.gov.za/publications/P0302/P03022011. pdf, accessed 24 April 2013), these percentages were different from the general South African population of age 10–14 years for whom 84.9% were self-identified black, 1.8% Indian/Asian, 5.4% white and 7.9% coloured, most likely due to the sample containing only urban, government schools. Most students reported that they were born in South Africa (94.9%) (data not shown in tables). Students’ self-reported skin colour was mostly light brown (54.3%), followed by white (21.5%), brown (15.1%), dark brown (5.9%) and black (2.6%). Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

53 67 96 113 42 66 92 91 87 707

7.5 9.5 13.6 15.9 5.9 9.4 13.0 12.9 12.3 100

Table 2. Demographic data of the students who completed the questionnaire Data Sex Male Female Missing Total Age 11 years old 12 years old 13 years old 14 years old or older Missing Total Ethnic group Black White Indian/Asian Coloured Don’t know/missing Total Skin colour White Light brown Brown Dark brown Black Missing Total

Number (%) 269 (38.1) 434 (61.4) 4 (0.5) 707 (100) 6 (0.8) 173 (24.5) 429 (60.7) 97 (13.7) 2 (0.3) 707 (100) 312 (39.8) 209 (26.7) 59 (7.5) 177 (22.6) 27 (3.4) 784 (100) 152 (21.5) 384 (54.3) 107 (15.1) 42 (5.9) 19 (2.6) 3 (0.4) 707 (100)

Note: Children were able to choose more than one ethnic group as many children have parents from different ethnic groups and consider themselves belonging to both groups.

269

Wright et al.

Table 3. Main effects of students’ sex on key outcomes for sun-related knowledge

Variable It is safe to get sunburnt once or twice a year †Things you can do to not get skin cancer: Avoid getting sunburnt Stay out of the summer sun Cover up with clothing Use sunscreen Get a suntan Go to a sunbed clinic Eat the rights foods Melanoma is a form of skin cancer Seen or heard about the UVI Heard about CANSA

All students (n = 707) % (n)

Male (n = 269) % (n)

Female (n = 434) % (n)

Sex OR

Sex difference P-value

95% CI

22.4 (158/705)

28.3 (76/269)

18.8 (81/432)

0.61

0.011*

0.41–0.89

64.5 (456) 48.0 (339) 45.5 (321) 65.4 (462) 5.2 (36) 14.9 (105) 38.0 (268) 13.3 (93/702)

64.6 (174) 45.3 (122) 48.3 (130) 63.9 (172) 8.9 (24) 15.8 (42) 37.5 (101) 13.0 (35/269)

64.2 (279) 50.0 (217) 43.7 (190) 66.5 (289) 3.0 (13) 14.7 (84) 38.2 (166) 13.3 (58/433)

1.02 0.82 1.20 0.89 3.17 1.07 0.97 1.0

0.915 0.231 0.240 0.473 0.000*** 0.755 0.852 0.992

0.74–1.39 0.61–1.13 0.88–1.63 0.65–1.22 1.58–6.38 0.7–1.63 0.71–1.33 0.63–1.58

28.8 (203) 74.4 (526)

32.7 (88/269) 71.5 (192)

26.4 (114/432) 78.6 (341)

0.74 1.81

0.072 0.008**

0.53–1.03 1.16–2.82

*P < 0.05; **P < 0.01; ***P < 0.001. Reference group: boys. The P-values for sex are for differences in adjusted means. CI, confidence interval; OR, odds ratio. Four students did not report their sex. †Categorical question where respondents could select as many option as they wish; however, there was no option to disagree with each category, therefore a lack of a positive response for a category was assumed as a negative or disagreement of the category, lack of agreement on all categories would be assumed as missing.

Students’ sun-related knowledge The results for the student’s sun-related knowledge responses are given in Tables 3 and 4. Many students (64.5%) believed that one could protect oneself from getting skin cancer by avoiding getting sunburnt. Other means reported to do so by the students were to use sunscreen (65.4%), stay out of the summer sun (48.0%), cover up with clothing (45.5%) and eat the right foods (38.0%). Only about a quarter of the students (22.4%) wrongly agreed that it is safe to get sunburnt once or twice a year. Few students (13.3%) had not heard about melanoma and knew that it was a form of skin cancer. It is probable that children had heard about skin cancer but had not heard specifically about melanoma and in particular, that it is a form of skin cancer. More than half of the students (55.4%) reported that they had not had any teaching about sun protection in the last 12 months of school. Despite this, some students (16.6%) believed that their school did have a sun protection policy. Only 28.8% of students had seen or heard about the UV Index (a measure of the sun’s intensity developed by the World Meteorological Organization) and for those students who had, some students (18.8%) had seen it on the television. Three-quarters of all students (74.4%) had heard about the Cancer Association of South Africa, which promotes sun awareness and skin cancer prevention. 270

Two comparisons are presented: one used boys as the reference group and the other used self-reported skin colours brown, dark brown and black combined. In the assessment of the influence of sex, for example, an odds ratio of > 1 suggests that girls were more likely to agree with a statement than boys, but less likely to do so if the odds ratio is < 1. For sun-related knowledge, there were only four questions with statistically significant differences for sex and skin colour. More boys (28.3%) than girls (18.8%) agreed that it is safe to get sunburnt once or twice a year (P = 0.011). More girls (78.6%) than boys (71.5%) reported that they had heard about CANSA (P = 0.008). More boys (8.9%) than girls (3.0%) reported that getting a suntan was a way to protect oneself from getting skin cancer (P < 0.000). More children reporting that they had white or light brown skin stated that melanoma was a form of skin cancer (15.1%) compared with children with brown, dark brown and black skin (7.7%) (P = 0.015). More children with white or light brown skin (68.4%) said that using sunscreen was a way to avoid getting skin cancer compared with children with brown, dark brown and black skin (55.9%) (P = 0.003).

Students’ sun-related attitudes Students’ sun-related attitudes responses are shown in Tables 5 and 6. Few of the students (8.7%) agreed that Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Sun exposure among South African children

Table 4. Main effects of students’ skin type on key outcomes for sun-related knowledge

Variable It is safe to get sunburnt once or twice a year †Things you can do to not get skin cancer: Avoid getting sunburnt Stay out of the summer sun Cover up with clothing Use sunscreen Get a suntan Go to a sunbed clinic Eat the rights foods Melanoma is a form of skin cancer Seen or heard about the UVI Heard about CANSA

All students (n = 707) % (n)

Skin colour brown, dark brown and black (n = 168) % (n)

Skin colour white and light brown Skin type (n = 536) difference % (n) Skin type OR P-value 95% CI

22.4 (158/705) 21.4 (36/168)

22.9 (122/534)

1.02

0.916

0.65–1.61

64.5 (456)

60.1 (101)

65.8 (353)

0.78

0.175

0.55–1.12

48.0 (339)

54.7 (92)

46.2 (248)

1.41

0.055

0.99–1.99

45.5 (321)

47.0 (79)

45.1 (242)

1.08

0.671

0.76–1.53

65.4 (462) 5.2 (36) 14.9 (105)

55.9 (94) 5.9 (10) 15.4 (26)

68.4 (367) 5.0 (27) 14.9 (80)

0.58 1.19 1.04

0.003** 0.643 0.861

0.41–0.84 0.56–2.52 0.65–1.69

38.0 (268) 13.3 (93/702)

52.8 (72) 7.7 (13/168)

36.3 (195) 15.1 (81/535)

1.31 2.13

0.130 0.015*

0.92–1.87 1.14–3.98

28.8 (203)

24.6 (41/167)

29.9 (160/535)

1.31

0.181

0.88–1.95

74.4 (526)

71.5 (118/165)

77.4 (407/526)

1.42

0.156

0.87–2.31

*P < 0.05; **P < 0.01; ***P < 0.001. Reference group: brown/dark brown/black. The P-values for skin colour are for differences in adjusted means. CI, confidence interval; OR, odds ratio. †Categorical question where respondents could select as many options as they wish; however, there was no option to disagree with each category, therefore a lack of a positive response for a category was assumed as a negative or disagreement of the category, lack of agreement on all categories would be assumed as missing.

they like to have a suntan because they feel healthier and that they think a suntan makes them feel more attractive to others (17.3%). Few also agreed that most of their friends (16.1%) and family (14.2%) think that a suntan is a good thing. Few students (9.9%) incorrectly agreed that tanned or dark skin protects against skin cancer. Regarding fashion, some students agreed that a suntan is less fashionable now than it used to be (20.2%) and that clothing that covers most of the arms and legs is not fashionable (23.1%). About a quarter of students (28.1%) believed that there was little chance that they would get skin cancer. More boys (13.4%) than girls (7.7%) agreed that tanned or dark skin protects you against skin cancer (P = 0.033). A strong statistical difference (P < 0.000) was found for children reportedly with white/light brown vs. brown/dark brown/black skin for the statement ‘I feel more attractive with a suntan’. More children with white/light brown skin

(20.8%) agreed with this statement compared with children with brown/dark brown/black skin (5.4%).

Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

271

Students’ sun-related behaviour As shown in Tables 7 and 8, approximately two-thirds (59.6%) of all students had got a suntan during the previous summer, but only 21.9% had purposefully sunbathed to try to get a suntan. More children reporting to have white/light brown skin (60.3%) stated that they had been sunburnt during the previous summer compared with children with brown/dark brown/black skin (41.7%) (P = 0.000). For sun protection, there were three statistically significant differences by skin colour groups. Children reporting to have white/light brown skin (88.4%) were more likely to agree that they stayed in the shade to protect themselves from getting sunburnt compared with children having

Wright et al.

Table 5. Main effects of students’ sex on key outcomes for sun-related attitudes

Variable Feel healthier with a suntan Feel more attractive with a suntan Friends think a suntan is a good thing Family think a suntan is a good thing Tanned or dark skin protects you against skin cancer Having a tan is less fashionable now than it used to be Clothing which covers most of the arms and legs is not fashionable There is little chance that I will get skin cancer

Sex OR

Sex difference P-value

95% CI

8.8 (38/433) 17.6 (76/431) 16.2 (70/128)

1.02 1.16 1.05

0.951 0.482 0.815

0.58–1.8 0.76–1.77 0.68–1.62

15.6 (42/269)

13.4 (58/434)

0.92

0.723

0.59–1.44

9.9 (69/699)

13.4 (36/269)

7.7 (33/430)

0.57

0.033*

0.34–0.96

20.2 (141/697)

23.1 (62/269)

18.5 (79/428)

0.76

0.187

0.5–1.14

23.1 (162/700)

24.3 (65/237)

22.4 (97/443)

0.91

0.626

0.63–1.32

28.1 (197/700)

25.5 (68/267)

29.8 (129/433)

1.35

0.127

0.92–1.98

All students % (n)

Male % (n)

Female % (n)

8.7 (60/701) 17.3 (121/700) 16.1 (113/700)

8.2 (22/268) 16.7 (35/269) 16.1 (43/267)

14.2 (100/703)

*P < 0.05; **P < 0.01; ***P < 0.001. Reference group: boys. The P-values for sex are for differences in adjusted means. CI, confidence interval; OR, odds ratio. Four students did not report their sex.

Table 6. Main effects of students’ skin type on key outcomes for sun-related attitudes

Variable Feel healthier with a suntan Feel more attractive with a suntan Friends think a suntan is a good thing Family think a suntan is a good thing Tanned or dark skin protects you against skin cancer Having a tan is less fashionable now than it used to be Clothing that covers most of the arms and legs is not fashionable There is little chance that I will get skin cancer

All students % (n)

Skin colour brown, Skin colour white Skin type dark brown and black and light brown Skin type difference % (n) % (n) OR P-value 95% CI

8.7 (60/701)

10.1 (17/168)

7.9 (42/534)

0.68

0.214

0.36–1.26

17.3 (121/700)

5.4 (9/167)

20.8 (111/534)

4.61

0.000***

2.22–9.59

16.1 (113/700) 14.4 (24/167)

16.5 (88/534)

1.1

0.704

0.66–1.84

14.2 (100/703) 15.5 (26/168)

13.6 (73/536)

0.83

0.468

0.5–1.38

9.6 (51/534)

0.86

0.625

0.48–1.55

20.2 (141/697) 19.3 (32/166)

20.5 (109/532)

1.11

0.680

0.69–1.78

23.1 (162/700) 23.8 (40/168)

22.9 (122/533)

0.98

0.929

0.64–1.5

28.1 (197/700) 21.0 (35/167)

30.2 (161/534)

1.6

0.042*

1.01–2.54

9.9 (69/699)

10.8 (18/166)

*P < 0.05; **P < 0.01; ***P < 0.001. Reference group: brown/dark brown/black. The P-values for skin colour are for differences in adjusted means. CI, confidence interval; OR, odds ratio.

272

Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Sun exposure among South African children

Table 7. Main effects of students’ sex on key outcomes for sun-related behaviour

Variable Got a suntan last summer Sunbathed regularly to get a suntan Did not use oils or lotions to get a suntan Did not apply a fake tanning lotion† Never used a sunbed Sunburnt last summer To protect yourself from getting sunburnt, did you ever: Stay inside Stay in the shade Wear a hat or cap Wear a broad-brimmed hat, bucket hat or cap with flaps Wear clothing (not a hat) Use sunscreen Applied a skin lightening cream

Sex OR

Sex difference P-value

95% CI

61.8 (267/432) 23.5 (102/434)

1.26 1.28

0.137 0.203

0.93–1.72 0.88–1.86

71.8 (193/269)

66.8 (290/434)

0.67

0.504

0.2–2.2

90.6 (663/701)

88.1 (236/268)

91.7 (397/433)

2.52

0.296

0.42–15.26

94.0 (659/701) 56.0 (393/703)

94.8 (254/268) 50.2 (135/269)

93.5 (405/433) 59.5 (258/434)

1.28 1.52

0.718 0.011

0.34–4.8 1.1–2.1

72.7 86.3 72.1 50.1

69.4 82.5 71.6 47.8

74.8 88.7 72.5 51.5

(323/432) (384/433) (313/432) (221/429)

1.31 1.66 1.04 1.16

0.122 0.021 0.816 0.335

0.93–1.83 1.07–2.56 0.74–1.46 0.86–1.58

55.2 (239/433) 68.6 (297/433) 20.9 (90/430)

1.13 1.36 1.44

0.419 0.062 0.08

0.84–1.54 0.98–1.87 0.96–2.16

All students % (n)

Male % (n)

Female % (n)

59.6 (418/701) 21.9 (154/702)

56.1 (151/269) 19.4 (52/268)

68.7 (483/703)

(509/700) (606/702) (505/700) (349/697)

53.4 (378/700) 66.0 (463/702) 18.9 (131/694)

(186/268) (222/269) (192/268) (128/268)

52.1 (139/267) 61.7 (166/269) 15.5 (41/264)

*P < 0.05; **P < 0.01; ***P < 0.001. Reference group: boys. The P-values for sex are for differences in adjusted means. CI, confidence interval; OR, odds ratio. Four students did not report their sex. †Variable had several categories; this was recoded into binary format for analysis as a result of low numbers in some cells.

brown/dark brown/black skin (79.8%) (P = 0.004). Similarly, children reporting to have white/light brown skin (69.4%) were more likely to agree that they used sunscreen to protect themselves from getting sunburnt compared with children having brown/dark brown/black skin (54.8%) (P = 0.0005). In contrast, children reporting to have brown, dark brown and black skin (61.1%) were more likely to agree that they wore clothing (not a hat) to protect themselves from getting sunburnt compared with children with white and light brown skin (51.9%) (P = 0.038).

DISCUSSION Generally, there was a mixed level of understanding about sun-related issues among the child respondents. Many children knew about CANSA, but few knew about melanoma skin cancer or had heard about the UV Index. While CANSA has been existence for several years working towards all types of cancer prevention, skin cancer prevention and awareness has not been extensive, no sun awareness programme exists in South Africa schools and the UVI is not broadcast to the public (it was for a short time in the late 1990s/early 2000s and was then abated). This suggests that more efforts are required Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

to raise the profile of sun-related issues in South Africa through media campaigns, and especially targeted at schoolchildren. Anecdotal evidence suggests that information on sun protection is presently lacking in the teaching curriculum, and this should be amended for the future. About 65% of children stated that they ever used sunscreen. This is a similar finding as found, for example, among New Zealand schoolchildren of comparable age of whom 51.4% used sunscreen (5). About 15% of South African schoolchildren who responded to the questionnaire said that they visited a sunbed clinic, a finding consistent with the 12.2% found among New Zealand schoolchildren of a similar age (5) but different from a United Kingdom study that found that 7% of primary school children had actually used a sunbed in the last 6 months (12). Sunbathing regularly last summer was affirmed by 22% of the South African schoolchildren, similar to 24.2% of New Zealand schoolchildren reported in Wright et al. (5). Few children (17.3%) felt more attractive with a suntan, but 59.4% had got a suntan during the previous summer. This is difficult to understand; however, it is probably that at least some of the children did not intend to get a suntan but received 273

Wright et al.

Table 8. Main effects of students’ skin type on key outcomes for sun-related behaviour

Variable Got a suntan last summer Sunbathed regularly to get a suntan Did not use oils or lotions to get a suntan Did not apply a fake tanning lotion† Never used a sunbed Sunburnt last summer To protect yourself from getting sunburnt, did you ever: Stay inside Stay in the shade Wear a hat or cap Wear a broadbrimmed hat, bucket hat or cap with flaps Wear clothing (not a hat) Use sunscreen Applied a skin lightening cream

All students % (n)

Skin colour brown, dark brown Skin colour white Skin type and black and light brown Skin type difference % (n) % (n) OR P-value 95% CI

59.6 (418/701) 53.5 (90/168)

61.2 (328/536)

1.36

0.078

0.96–1.94

21.9 (154/702) 23.9 (40/168)

21.2 (113/536)

0.85

0.464

0.56–1.29

68.7 (483/703) 72.0 (120/168)

67.6 (362/536)

0.25

0.152

0.03–1.95

90.6 (663/701) 88.10 (148/168)

91.0 (486/534)

1.37

0.265

0.79–2.38

94.0 (659/701) 92.9 (156/168) 56.0 (393/703) 41.7 (70/168)

94.4 (504/534) 60.3 (323/536)

0.4 2.27

0.379 0.000***

0.05–3.26 1.56–3.29

72.7 86.3 72.1 50.1

72.5 88.4 71.6 49.6

(388/535) (472/534) (384/536) (264/532)

0.97 1.93 0.86 0.94

0.893 0.004** 0.466 0.722

0.66–1.44 1.22–3.07 0.58–1.28 0.66–1.33

53.4 (378/700) 61.1 (102/167)

51.9 (277/534)

0.69

0.038*

0.48–0.98

66.0 (463/702) 54.8 (92/168) 18.9 (131/694) 22.9 (38/166)

69.4 (371/535) 17.6 (93/529)

1.87 0.72

0.0005** 0.127

1.31–2.67 0.47–1.1

(509/700) (606/702) (505/700) (349/697)

73.1 79.8 74.6 51.2

(122/167) (134/168) (123/165) (85/166)

*P < 0.05; **P < 0.01; ***P < 0.001. Reference group: Brown/dark brown/black. The P values for skin colour are for differences in adjusted means. CI, confidence interval; OR, odds ratio. †Variable had several categories; this was recoded into binary format for analysis as a result of low numbers in some cells.

accidental sun exposure and their skin darkened, leading to tanned skin without the desire for tanned skin being in place. It is acknowledged that further research is necessary to validate and fully understand these findings in the South African context. While the CANSA’s efforts to promote sun protection through the use of sunscreen, shade, sunglasses and clothing, as indicated on their web site and through flyers distributed at their awareness campaign events, the present study is unable to evaluate the impact of these efforts on South African primary schoolchildren. While few children agreed (22.4%) that it is safe to get sunburnt once or twice a year, with statistically fewer females in agreement than males, almost 60% of children reported that they had got a suntan. This is comparable with a similar international study where 61.8% (5) of children in New Zealand reported that they got a suntan. Similarly, 56% of children reported that they had been

sunburnt last summer, a finding consistent with Swiss (13) and New Zealand (5) schoolchildren. Children either did not know how best to protect themselves from getting sunburnt, did not place sufficient value on their health to change their behaviour or did not experience enough support from parents or caregivers to make use of sun protective practices and behaviour. In a South African context, more in-depth discussions may be necessary with children of different skin types to ascertain their understanding of sunburn and suntan, as this may skew their responses influenced by individual perception of risk according to individual skin colour. The number of sunburns experienced during the previous summer should also be assessed. Evidence has shown that a reasonable level of sunrelated knowledge, especially regarding skin cancer, does not necessarily translate into skin cancer prevention behaviours (4) and that the relationship between sun-related

274

Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Sun exposure among South African children

knowledge and behaviours may be mediated by attitudes (5). This has not yet been tested among South African schoolchildren; however, here we found that few children had pro-tan sun attitudes based on the eight attitude variables, which may have a positive influence on their sunrelated behaviours and reduce the risk of sun exposure as they prefer not to suntan. For these questions, there was a possible answer of ‘Be unsure’ and many students answered these questions with this response. Future studies should eliminate this response option, or find an alternate way for better question comprehension to improve completion rates. As expected, more children with white/light brown skin agreed that having a suntan makes them feel more attractive, compared with children with brown/dark brown/ black skin. It may be that children with white/light brown skin are more influenced by social pressures, i.e. in the media, encouraging one to desire a suntan and darker skin, compared with children of other skin types. Children with darker skin may also not care for tanned skin; however, this should be further researched in the South African context. Furthermore, this study suggested that reported behaviour (at least for sunburnt last summer and for some types of sun protection used) may sometimes be influenced by skin sensitivity. Future research is planned to fully analyse the relationships between reported risky behaviour and knowledge among South African ethnic groups. In general, the number of children reporting the use of different sun protection options to protect one from getting sunburnt was promising. Many children agreed that they sometimes, most of the time and/or always stayed inside or in the shade, wore a hat or cap and sunscreen. Fewer children agreed that they used a broad-brimmed hat, bucket hat or cap with flaps, which is unfortunate as these types of hat are more protective than a cap, for example. Similar findings have been made in other studies (5, 7). Among Australian children, hats worn most by children were peaked caps (70%), legionnaire caps (46%), wide-brimmed hats (18%) and narrow-brimmed hats (15%) – the total adds to more than 100% because some children wore more than one hat type (14). The use of clothing (other than a hat) for sun protection was not a preferred choice among the children, and this may be influenced by school uniform restrictions and fashion choices. Statistically more children with white/light brown skin reportedly used sunscreen compared with children of other skin colours as one would expect given the natural protection melanin affords individuals with dark skin. This may be because they, or their parents and caregivers, are

more aware of their inherent susceptibility to burn and that suitable products are available on the market for them which may not be the case for children with brown/dark brown/black skin. While there were few statistically significant differences in sun-related knowledge, attitudes and behaviours between children with brown/dark brown/ black skin and children with white/light brown skin, differences do exist and these differences should be considered when developing targeted skin cancer prevention and sun awareness programmes for South Africa. Furthermore, previous studies (14) have found age differences for sun-related knowledge, attitudes and behaviours, and this should also be investigated in the South African context across grade levels. Some possible study limitations include the reliance on the self-reporting of usual behaviour that is strongly contextually linked. However, previous studies have shown that children are reasonably good at reporting, including their sun protection behaviour (15). Children’s understanding of certain questionnaire items is also of some concern, for example, sunbed use appeared to be more frequent among individuals with darker skin, which is questionable. Focus group discussions among children aged 11–13 years may help gauge whether children know what a sunbed is and whether such use is in existence. The study was carried out during the third school term (late winter, early spring), and this may have an influence on the children’s responses compared to had the questionnaire been carried out during summer months. The reason for the questionnaire being administered in the third term was a condition of the government approvals given for the study as part of the ethical clearance process. Questionnaires were sent via a postal survey, and class lists were not provided to the researchers, therefore it was not possible to determine number of nonresponders. Exclusion criteria applied means that the survey results are not representative of the whole South African primary school population but instead only government, urban schoolchildren in the nine provinces. The study sample was not representative of the South African population aged 11–13 years in terms of skin type; there were only 40% black children compared with 80% black children in the entire South African population of this age. Although schools were randomly selected from the government urban primary school population in all nine provinces, exclusion of rural schools may have contributed towards this skewed representivity. Hence, these study results may help inform the design of intervention programmes, but more detailed research is required among children with darker skin to prepare appropriate health promotion messages accordingly.

Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

275

Wright et al.

CONCLUSIONS The most important insights gained in this study were that South African schoolchildren at urban government schools do have some knowledge about sun protection, and they do have some positive sun behaviours; however, the reported occurrence of sunburn, a risk factor for skin cancer, was relatively high. This was also the case for the occurrence of suntanning that has been associated with excess solar UVR exposure and sunburn. There is evidence suggesting that public health approaches to changing people’s health behaviours has effect but that a ‘one size fits all’ approach to skin cancer prevention is not effective and tailored campaigns and interventions have more success (4). While it is apparent that there is a need to increase sun-related knowledge among all ethnic groups, a focus on children with fair skin is important given their higherrisk of skin cancer compared to children with darker skin tones. Here, few statistically significant differences were found between sex and skin colour and children’s sunrelated knowledge, attitudes and behaviours. Future research should consider age differences and use an objective measure of ethnicity to compare different ethnic groups. Despite several shortcomings, these

findings are of relevance for the formation of appropriate sun protection programmes aimed at schoolchildren in South Africa and other countries with similar multiethnic populations.

LIMITATIONS A limitation of this study is the use of self-reported questionnaires, completed by primary schoolchildren, to assess their sun-related knowledge, attitudes and behaviours.

ACKNOWLEDGEMENTS Funding for this project was granted by the Cancer Association of South Africa, the Council for Scientific and Industrial Research and the Medical Research Council of South Africa. Sincere thanks go to the School Principals, School Teachers and Schoolchildren who participated in the study and completed the questionnaires. The authors would like to thank Research Associate Professor Anthony Reeder for reading an early version of this manuscript and providing valuable recommendations for its improvement.

REFERENCES 1. IARC. Solar and ultraviolet radiation. IARC Monogr Eval Carcinog Risks Hum 1992; 55: 1–227. 2. Community Preventive Services Task Force. Preventing skin cancer: primary and middle school interventions. 2012. Available at http://www.thecommunity guide.org/cancer/skin/education-policy/ primaryandmiddleschools.html. Accessed 27 January 2014. 3. Lynagh M, Schofield M, Sanson-Fisher R. School health promotion programs over the past decade: a review of the smoking, alcohol and solar protection literature. Health Promot Int 1997; 12: 43–60. 4. Keeney S, McKenna H, Fleming P, McIlfatrick S. Attitudes, knowledge and behaviours with regard to skin cancer: a literature review. Eur J Oncol Nurs 2009; 13: 29–35. 5. Wright C, Reeder A, Gray A, Cox B. Child sun protection: attitudes mediate the association between children’s knowledge and behaviours. J Paediatr Child Health 2008; 44: 692–698.

276

6. Ergul S, Ozeren E. Sun protection behaviour and individual risk factors of Turkish Primary School Students associated with skin cancer: a questionnaire-based study. Asian Pac J Cancer Prev 2011; 12: 765–770. 7. Tempark T, Chatproedprai S, Wananukul S. Attitudes, knowledge and behaviours of secondary school adolescents regarding protection from sun exposure: a survey in Bangkok, Thailand. Photodermatol Photoimmunol Photomed 2012; 28: 200–206. 8. Mqoqi N, Kellet P, Sitas F, Jula M. Incidence of histologically diagnosed cancer in South Africa, 1998–1999. Pretoria, South Africa: National Cancer Registry, 2004. 9. Ncongwane KP, Coetzee GJR. A quantitative analysis of the SAWS GAW UVB network and data since 1994. South African Society for Atmospheric Science Conference Proceedings. 2010. 22–23 September 2010. ISBN 978-0-62050849-0. 10. Jopson J, Reeder AI. Sun protection in New Zealand secondary schools: obstacles and opportunities. Dunedin: Social and

11.

12.

13.

14.

15.

Behavioural Research in Cancer Group, Department of Preventive and Social Medicine, Dunedin School of Medicine, University of Otago, 2004. Bulliard J-L, Reeder AI. Getting the message across: sun protection information in media weather reports in New Zealand. N Z Med J 2001; 114: 67–70. Hamlet N, Kennedy K. Reconnaissance study of sunbed use by primary school children in Lanarkshire. J Public Health 2004; 26: 31–33. Reinau D, Meier C, Gerber N, Hofbauer GF, Surber C. Sun protective behaviour of primary and secondary school students in North-Western Switzerland. Swiss Med Wkly 2012; 142: w13520. Dixon H, Borland R, Hill D. Sun protection and sunburn in primary school children: the influence of age, gender and coloring. Prev Med 1999; 28: 119–130. Riley AW. Evidence that school-age children can self-report on their health. Ambul Pediatr 2004; 4 (4 Suppl): 371– 376.

Photodermatol Photoimmunol Photomed 2014; 30: 266–276 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Self-reported sun-related knowledge, attitudes and behaviours among schoolchildren attending South African primary schools.

Skin cancer and other adverse health effects result from excess solar ultraviolet radiation exposure. Sun protective practices are important intervent...
91KB Sizes 0 Downloads 0 Views