Recreational exposure to the sun may not explain fully current trends
in melanoma incidence. The hypothesis was examined whether carcinogens
in water play a role in the development of cutaneous melanoma. In a c
ase-control study, 128 melanoma patients and 168 patients with other t
ypes of malignancy completed a detailed questionnaire on aquatic leisu
re time activities. All relative risk estimates were adjusted for age,
gender, educational level, pigmentation characteristics, and exposure
to sun habits. Regular swimming during the summer months in swimming
pools and in open waters such as rivers and seas before the age of 15
years, was associated with odds ratios of 2.20 (95% confidence interva
l (Cl), 1.05-4.62) and 2.41 (95% Cl, 1.04-5.58), respectively, compare
d with no swimming at all or swimming in relatively unpolluted waters,
such as lakes and fens. Melanoma patients learned to swim at a younge
r age; compared with those who never learned to swim or who learned to
swim after the age of 12 years, the odds ratio was 1.87 (95% Cl, 0.91
-3.78) for those who learned to swim at ages 9-12 years, and 2.22 (95%
Cl, 1.16-4.26) for those who learned to swim before 9 years of age. C
ompared with persons who had no swimming certificates, an odds ratio o
f 1.25 (95% Cl, 0.71-2.23) was found for persons with one or two certi
ficates, and an odds ratio of 2.96 (95% Cl, 1.25-6.96) for persons wit
h three or more certificates. The positive association between a histo
ry of swimming and melanoma risk suggests that carcinogenic agents in
water, possibly chlorination by products, play a role in melanoma aeti
ology.