The subject of this study was to investigate the relationship between prescription of cholesterol-lowering drugs and depression. We used prescription of antidepressants as a proxy for depression and analysed the prescription order for cholesterol-lowering and antidepressant drugs. The ratio of persons with antidepressants prescribed second and ®rst translates directly into a rate ratio (RR) associating cholesterol-lowering drugs use with antidepressants. The crude RRs were then adjusted for trends in sales of the drugs over the study period. All residents of Funen, Denmark, who started the two therapies during the period 1 April 1991 through 31 December 1995 were included in the study cohort. Of 184 individuals included in the analysis, 105 started antidepressant ®rst and 79 second, giving a crude RR of 0.75. However, the sales of cholesterol-lowering drugs increased more than the sales of antidepressants. Accordingly, the adjusted RR was higher than the crude, 0.90 (95% con®dence interval 0.68 to 1.22). Among three tested cholesterol-lowering drug classes, only simvastatin showed an adjusted RR above unity (1.59, 95% con®dence interval 1.08 to 2.45). The hypothesis that the use of cholesterol-lowering drugs has an adverse eect on mood is not supported by the present study. Confounding by indication might explain the apparent association between use of simvastatin and antidepressants.