PURPOSE: To examine the association between perceived stress and use of over-the-counter analgesics in a representative sample of 25-44-year old adults, and to examine the association across various socio-demographic strata. Furthermore, to examine whether an association between perceived stress and use of over-the-counter analgesics attenuates when controlled by potential stress-related pain and discomfort. METHODS: National representative cross-sectional study in Denmark. The study population consisted of men and women aged 25-44 years, n = 4739. The survey was conducted by face-to-face interviews. The outcome measure was use of over-the-counter analgesics (OTCA). The independent variable was perceived stress. Demographic variables and pain and discomfort symptoms were included as covariates. RESULTS: Analyses stratified by socio-demographic factors (gender, education, cohabiting status and whether or not the respondents had children) all showed a significant and graded association between stress and OTCA use. The odds for OTCA use mounted with increasing stress. In analyses adjusted for socio-demographic variables and pain or discomfort the association between stress and OTCA use attenuated somewhat, but remained significant and graded. The crude odds ratio (OR) for OTCA use was 1.36 (1.19-1.55) among participants who sometimes felt stress, and 1.91 (1.58-2.30) among participants who often felt stress, compared to participants without stress. CONCLUSION: There was a significant and graded association between perceived stress and OTCA use. The association was robust across all the examined socio-demographic strata and could not be explained by potential stress-related pain and discomfort. The results indicate that OTCA are used inappropriately to treat feelings of stress. Copyright (c) 2009 John Wiley & Sons, Ltd.